software.dev

Introducing The Social Postal Service!
http://apps.facebook.com/socialpostal

The idea behind the Social Postal Service is simple - send free, physical cards to your friends for various holidays - starting with Valentines Day! Pick from 15 designs, type a message, an address, and you’re done. It’s that simple. Why poke friends on Facebook when you can ‘poke’ them in real-life with a real card?

Each card costs ‘Card Points’, a unit of virtual currency that users earn by completing surveys from our advertisers. These surveys help to pay the bill, as it were, and makes it possible to send a physical product to our users.

The Social Postal Service also gives users the ability to send virtual ‘Conversation Hearts’ to your friends. These are those popular, colorful sugar hearts that always come out around Valentines Day with little sayings on them. You can send virtual versions of these hearts to your friends, customized with your own message.

Give the Social Postal Service a try and let me know what you think!

http://apps.facebook.com/socialpostal

Another day, another app!

Where Should We Eat?
This one is called, ‘Where Should We Eat?’.  The basic premise is to address the problem of just that.  How many times have you gone out to eat with friends & family, and no one wanted to pick a place.  With the click of a button, you can easily find a random restaurant nearby.

I got to use the Yelp API and JSON for this, neither of which I’ve used before.  JSON was super easy to work with, thanks to PHP’s handy “json_decode()” function.

I doubt this app will get much adoption since you don’t need to add it like most apps, and it doesn’t have a viral component, but it was quick and easy to make, and actually pretty useful.  I’ve gotten positive feedback on it so far — give it a try!

Try the Mad-Libber on Facebook!

Been a LONG time since I wrote in this thing (nearly a year!) but I certainly haven’t been idle on the coding front.

Recently, I’ve been particularly interested building Facebook applications. It’s really an easy platform to develop for. Facebook provides great APIs and documentation, and you can bring pretty much whichever language you want to the table. The forums are also very active, with lots of helpful posters.

My first application, the Mad-Libber, is basically a Facebook version of the popular Mad-Libs game. For those of you who don’t know what that is, Mad-Libs are stories with blanks for certain words. The reader is given the type of word (noun, verb, adjective, etc… ) but not the context. The idea is that the reader supplies these words for the story and you read the result, obviously making no sense because the context isn’t known to the reader. That’s what makes it funny.

With my application, users can write these stories, complete stories their friends wrote, and share them with friends. I’m going to try and monetize this through ads, which is actually quite successful for many applications. Cubics, one of two ad networks I use (the other being Appsaholic) offers $0.50 per thousand impressions. If you could get 50,000 page views a day, that’d be about $25 /day or $750 a month. For an independent developer who does this as a hobby, that’s not too bad.

I’ve found that it can be really difficult to get traction for your application, however. A viral nature is a must. You look at the top apps for Facebook, like SuperPoke, Graffiti, etc… and all of them have a very viral component which makes them easy and fun for people to spread. Another thing to keep in mind is that it doesn’t matter which application comes out first, or which one is ‘better’, but what matters is that yours catches on.

If you have some time, give my application a try at: http://apps.facebook.com/madlibber. If any of you are interested in developing Facebook applications, I’d be glad to share my experiences and provide some advice. Just post here or email directly at ejfarraro@ucdavis.edu.

Thanks and happy coding

*** Some people mentioned they were able to access the account over the phone with Dell without any other information, so I’ve removed all the numbers from the letter my coworker received.
* Short break from Software Dev posts — I wanted to get this out there.

I sure do. I remember there was once a time when the Dell name was associated with quality computers. Apparently, they’ve taken the wrong path somewhere down the line, and my experience with Dell has been less then stellar as of late. Case in point, I’d like to tell the story of what happened to my coworker.

My coworker recently had problems with his Dell laptop and sent it in. Not too surprising — laptops move around a lot, thinks might come loose, stuff stops working. It happens. I’ve worked with many laptops, and never had one not have some sort of problem after 2-3 years. Needless to say, it makes sense to buy a warranty for your laptop when you buy one because these things tend to break. My coworker is fully aware of this, and purchased some sort of extended warranty plan for his Dell laptop.

So after sending it in and waiting several weeks, he began to wonder when he was going to get his laptop back. He talked to Dell service representitives initially by phone, and though they weren’t able to provide him with any information, they assured him they would call him back when more details were available. Unfortunately, the call never came. My coworker decided to try the online chatroom help instead, since he didn’t seem to be getting the help he needed. The online service chats were no more useful.

At this point, over a month has passed since the laptop was mailed away, and he hasn’t been able to get any sort of straight answer. I observed him chatting with the service representatives online, and frankly, the level of service was ridiculous. My coworker asked when he would get his laptop — the representative responded with things like “You should have more information in 72 hours”. 72 hours just to get more information about the status of a laptop that was sent in weeks ago?! The representative also assured my coworker that if for some reason his laptop couldn’t be fixed, a new one would be sent to him.

Today, my coworker received his laptop back from Dell much to his suprise. Finally, the wait was over! Or was it? Inside there was a note, describing the repairs done. This is what it said:

Talk about a slap in the face. Not only does it still not work, the ‘Thanks!’ at the bottom is like a cruel taunt.

I’m writing this in hopes that someone at Dell sees it, and rectifies this example of poor customer service.

I forgot to post about the gadgets after the contest deadline was over!

I ended up submitting two gadgets — one I called ‘Hangman 2.0′, in two Web 2.0 spirit. The other I called ‘Flippr’; it’s a flash-card app that allows you to add / remove cards, and then test your knowledge. Saving is supported by exported XML.

Flippr

Hangman 2.0
You can give the Gadgets a try if you’d like with the links above.

Hangman 2.0: This was the first gadget I did. The idea was that I often find the online Hangman apps boring. Usually they use super intellectual words or categories I don’t care about. Wouldn’t it be cool, I thought, if *I* could pick the category, and get a random word from it? And what better way to accomplish this then Google. I tried a few techniques for generating the word list, but I eventually settled on using Google to find the appropriate Wikipedia entry for a category, and scraping the entry for words. I sorted the words based on their frequency, and after removing a bunch of common words like ‘the’, ‘and’, etc… I randomly pick one of the top 30 or so. The result is pretty good; occasionally you’ll get a poor word, but most of the time it’s pretty fun. I also did some original pixel art to add a little flavor to the game, instead of the traditional hangman’s noose.

Flippr: I worked on this gadget second. I’ve never really had a chance to use flash cards for any class because in most engineering classes, flash cards wouldn’t help. However, I’m taking an econ class this quarter for GE credit, so I decided flash cards might help out. Flippr is pretty simple: basically you create a new ‘deck’, and then you can add, view, and remove cards from your deck. Once you’re finished with that, there is a ‘test’ option, where you can go through your deck, read each card and then FLIP (hence the name) to see if you know what the term or idea means. All in all, I’m happy with this turned out, though I wasn’t able to get the XML loading working in IE.

On a sad note, I got the Google rejection letter five days after I applied without a phone screen; if I wasn’t going to make it, I was hoping to get cut at AT LEAST the phone screen; but so much for that Then today I got another rejection, this time from Microsoft (I got cut after an on-campus interview). My GPA (3.4ish) isn’t horrible, but if I end up flipping burgers, I’ll have a long time to think about what happened.

Life is good!

I was actually pretty surprised how fast the news propagated through the internet. After posting the details of the exploit on Digg at about 8pm tonight, it drifted off into obscurity. The following morning however, apparently a bunch of people some how found the article, because it was already on the frontpage with 300 Diggs. Several hours later, the details were ALL over — Slashdot, Newsforge, Digg — all the big sites had discussions about the exploit. In some cases, people were confused and I did my best to link back to the article so people knew what was going on.

It wasn’t clear to everyone from my previous article, but Google knew of the exploit several weeks ago and they immediately shutdown the login page for the service. While existing pages were left up, they did the right thing immediately after I reported to them the details of the bug by making sure no one could actually exploit it.

While I believe I took appropriate action by first disclosing the exploit to Google and waiting until the service was closed for several weeks, there were a few people concerned with the disclosure. It is in my opinion that leaving the exploit quiet is a far worse scenario — while I don’t know of any other sites like this in existence, I think a headline of this nature is FAR preferable to a headline that says ‘Thousands give credit card numbers in Google Phising exploit’. Thousands of people were able to learn about what the whole ‘http://www.google.com/u/’ style URL meant, and will not fall for this exploit if it exists in the wild somewhere.

As best I can tell, I served about 245,000 requests and served 20 GB of pictures.  I’m happy to say that my host, Dreamhost, prevented the site from going down during the Digg/Slashdot/etc… rush!

Thanks to everyone who helped spread the word and left comments. Keep and eye on this site as well — I definitely love looking for these kinds of things

You can read the official report by Google here:

http://googlewebmastercentral.blogspot.com/2006/09/for-those-wondering-about-public.html