Paul Tarjan's Blog

Facebook Registration Tool

2010-12-16T00:00:00-08:00

Republished from Facebook Developer’s Blog

Today we’re launching a new registration tool that gives website owners the ability to offer quick, easy and social options for sign-up.

It is an excellent alternative to using Facebook Login (formerly Facebook Connect) when:

you want to provide an option for users without Facebook accounts
your site needs additional information that Facebook doesn’t provide
a traditional HTML form suits your site more

By minimizing the friction associated with signing up for a new account and making it easy for people to bring their friends with them, we’ve seen that people are more likely to complete the sign up process, stay on sites longer, share more content, and come back more often. For example, in beta tests with FriendFeed, Facebook sign ups increased 300%.

How it works

The registration tool is an iframe that websites can add with just one line of code, and customize to request the specific fields required to create an account. When a user is logged into Facebook and arrives at the registration tool on a website, they’ll see that the form is prefilled with the relevant information he or she has already shared on their Facebook profile. Users can see the specific information the site is requesting of them, giving them more control to decide whether to sign up.

The site can request information stored on Facebook (e.g., name, verified email address, current location, etc.) or custom information (e.g., text fields, checkboxes, dropdown menus, security checks, typeaheads, etc.). Users can decide whether to share their information with the site by clicking “Register,” or if a Facebook account isn’t required by the site, select an option to fill out the form manually. Because the tool works through an iframe, no data is shared with the website until the user clicks “Register.”

When a user registers for the site with their Facebook account (similar to Facebook Login), the site is given permission to access the user’s basic information on Facebook in addition to the data presented in the form, and in turn can provide the user with a personalized and social experience.

Getting started

To integrate the registration tool, simply add the iframe or XFBML to your site:

Using iframe:

<iframe src="http://www.facebook.com/plugins/registration.php?
             client_id=113869198637480&
             redirect_uri=http%3A%2F%2Fdevelopers.facebook.com%2Ftools%2Fecho%2F&
             fields=name,birthday,gender,location,email" 
        scrolling="auto" 
        frameborder="no" 
        style="border:none;" 
        allowTransparency="true"
        width="100%" 
        height="310px">
</iframe>

The redirect-uri should point to the page which will process your signed_request. You can customize the fields attribute to suit your needs. See the docs for a list of valid fields.

Using XFBML:

<fb:registration 
  fields="name,birthday,gender,location,email" 
  redirect-uri="http://developers.facebook.com/tools/echo/">
</fb:registration>

The redirect-uri and fields are identical to above. You must use your own app_id for the redirect-uri to be on your domain.

For more details on getting started with your implementation, please visit our technical documentation.

We hope you’ll try it out and let us know what you think.

Paul, an engineer on the Platform team, will register for your site if you use this tool. 1 more user!

Wisconsin Hackathon 2010

2010-09-28T00:00:00-07:00

Our second Camp Hackathon was at the University of Wisconsin. Amidst all the wonderful cheese and midwest manners, we found a bunch of great hackers. This time, it was a 12 hour hackathon so every single one of the Facebook engineers stayed up the whole time! No cheating and swapping out for a little shut-eye. Even the videographer from the student paper stayed the whole time, filming away for a wonderful front-page article they did on our little Hackathon.

Best Product Idea - REX

This won the “If you want to drop out of school – not that we recommend that at all – this could work” category. It recommends music based on what your social network is listening to.

Best Use of the Keyboard - NewsFeed Navigation

A greasemonkey script that elegantly lets you skim around your News Feed, like, comment, and then keep browsing, all without a mouse.

Best Totally Crazy Hack - Face Battle

A Java app, that takes a photo of your face, converts it to map tiles (desert, grassland, water, etc) and then plays a tank battle across your face.

We didn’t get a screenshot, so here is an action shot of his hacking setup.

Best Non-Crazy Solo Hack - Track your IMDb 250

This web application allows you to manage and watch all the top 250 movies on the IMDb 250 list. This list represents the best movies made in the last 100 years and is a great start for movie enthusiasts and movie buffs like me. This app crawls through IMDb, fetches the list and other metadata, and allows the user to automatically add movies to their Netflix disc queue. This feature was implemented using Netflix RESTful API.

Most Practical - Classmate Finder

A Facebook app that allows users to connect with their classmates in Facebook. The app requires the user to upload their class schedule and will then present the user with a list of facebook users for each of their classes.

Technically Hardest Problem - Video Face Tracking

Instead of tagging people in pictures, why can’t you tag them in segments of images? This hack does facial detection in various segments and then prompts you tag your friends in every part of the movie. It was a little rough around the edges (as all hacks should be), but has promise.

Best Mashup - Hola Mundo

It’s a map-based application that mashes Google Maps API, Google’s Translate API, Flickr API, and Yahoo News’ API to create a friendly postcard.

First Prize - Magic Grade Ball

UW-Madison publishes the summary statistics of grades for every class. This hack parses in that data, and if you put in your previous grades, it can predict your future grade. I can’t wait to see if it works for these students :)

Grand Prize - SoundStroke

A program that lets you draw on a canvas and have sound produced as you draw. The sounds you get are based on the shape of the strokes you drew, and the different colors are mixed into a “sound loop”. A drum machine cross with MSPaint.

Thanks

Thank you to all of our amazing engineers Tony Kamenick, Venkat Venkataramani, Arun Vijayvergiya, and awesome recruiter Marcia Velencia.

Keep Hacking!
Paul Tarjan

GATech Hackathon 2010

2010-09-14T00:00:00-07:00

We just finished our first ever Camp Hackathon at Georgia Tech. I think this is the most people I’ve ever seen stay up for the whole 24 hours. Our very own Justin Mitchell pulled the all-nighter, and then continued on until 3am the next morning! I haven’t heard from him, I hope he’s still alive…

Here is the answer to “who stayed up all night?”

Viral Award - Scategories

As a fun twist, we said whoever could get the most likes on their post, would win a prize. After only about 50 minutes, the winners had a whopping 37 likes! What a friend-graph they have.

They built a scategories Java app based on the course descriptions for their 8 specializations.

Best Video Sharing App - Virtual and Social Collaborative Video

A website that lets you synchronize watching videos with your friends in real-time. It was built using HTML 5 and Comet.

Best Birthday Wishing Service - Lazy Wisher

Taking slacking to whole new heights! This app will automatically ease the pain of having over 500+ ‘friends’. Just add this app and let the automatic b’day wishes flow from your ‘heart’ and into your ‘friends’ wall. Be the ‘friend’ everyone thinks you are and install this app.

Second Place - Social Juke Box

Social Juke Box is an application that lets you browse through music suggested by your friends to you. In the author’s opinion, this is what Apple Ping should have been.

First Place - DroiStick

A pair of handheld game controllers for multi player Facebook Games. This particular project implemented a ping pong game in JavaScript as a Facebook application which communicates with the clients (Android handheld device). All the interaction and motion of the game is dependent on the motion of the handheld device (similar to Wii controllers).

Grand Prize - FB-Lingo

A simple greasemonkey script built ontop of Google Transliterate, that lets you type in English but post the characters in any language you want. In addition, you can translation any languages characters into their closest English sounds.

Thanks

Thank you to all of our amazing engineers Hisham Harik, Justin Mitchell, Kiran Sidhu, and awesome recruiter Clifton Tay.

Keep Hacking!
Paul Tarjan

Debugging URLs with the Facebook URL Linter

2010-06-10T00:00:00-07:00

Republished from Facebook Developer’s Blog

A primary design goal of the Open Graph protocol – which enables any Web page to become a rich object in the social graph – has been to make it simple for developers to integrate it into their Web pages. But, as with any coding, syntax errors can and do occur. To help you debug your URLs, we’ve released the Facebook URL Linter.

You can use the Facebook URL Linter with many types of URLs, in addition to Open Graph protocol pages. Try the linter on our homegrown examples or on some pages in the wild.

At this time, the linter returns warnings for any issues it encounters. In the near future, the linter will start returning errors. These errors will cause the Like button to stop working on your page.

We also plan to add Open Graph protocol type metadata suggestions, and a lot of helpful messaging. If you have any feedback, we’d love to hear it on the Developer Forum. Make sure to put linter somewhere in the post, to help us find it.

This week we’ve also added Open Graph protocol markup to every public Page on Facebook. This makes it easy for anyone to know that Starbucks is a company, the White House is part of the government, and Weird Al is my favorite musician.

Lint away!

Paul is a lint-free, white and nerdy hacker on the Platform team.

Note: In case you were wondering, the Facebook URL Linter’s name hails from the original lint UNIX program.

Slogging

2010-05-29T00:00:00-07:00

I like to follow Nike’s law of hacking: Just Do It. Sadly, I’m a little tight on time right now and want to get some feedback on this idea before I actually build it.

I really like Wikiepdia’s model for articles. A template with some cool facts and a few sections for longer text. Often, I want to write posts and updates like this, but various excuses always crop up. I forget what I did last time, it doesn’t exactly fit my new post, I get lazy and stop posting, etc.

Sometimes I write one-off articles, but most times my posts will be similar. Video game reviews, product announcements, trip photos, etc. I wish I had a nice way to define templates and fill them in. Sure, that’s not too hard, there are billions of template engines.

The really cool part would be when different people use the same templates. Then you can cross reference them and building a micro site around those templates. One such template is Yelp reviews; another is Quora questions. You could build up which templates are most used, which keys in which templates are most filled in, which values are allowed in which fields, and grow these templates organically. So when I sit down to write about Super Mario Galaxy 2, there is already a high quality community template for video game reviews where I put in the game title, my hours played, my dates played, enjoyable rating, and a little blurb on how I felt. I don’t have to think about what would be good facts to write down, someone smarter than me already did that. Then an agregator can look through all the reviews and show an interesting view from all posters.

I pitched this idea to my friend Yury and we thought the core of this system is Structured bLOGing. Thus, it should get the abhorrent name slogging.

There are still many unanswered questions, and the exact implementation details are a bit muddy. So, I ask my friends and readers a few things:

Do you want to write structured posts too? Do you want to host them yourself or would a central site be good?
Is there existing software that could get me 90% the way there? Wikis? Template engines?
Does a UI ninja have any neat ideas? I want it to be a little more than a form with a bunch of <textarea>s.
Again for a data mining ninja. What neat things would you do with a bunch of structured posts?
Can someone PLEASE come up with a better name? I hate the name blogging and I hate slogging more. (and the .com is taken)
Anything else on your mind?

Graph Likes Over Time

2010-05-03T00:00:00-07:00

Wouldn't it be nice to see how many likes a page gets over time? Actually, it's really simple with webnumbr:

Enter the url you want to track into the box below
Click 'Create Numbr' on webnumbr
Watch your pretty graph

Here is all the current like graphs.

Don't Throw Your Money Away

2010-05-02T00:00:00-07:00

If you are ever buying something from Amazon, I can get 4% of your whole purchase as a ‘referral fee’. If you let me know what you bought and your paypal, I’m more than happy to send you the money, or I’ll buy you a drink in your honour.

If you just start your shopping with this link:

Paul’s amazon.com link

I get the referral as long as you buy your item before you close your web browser (once you click that link).

If you don’t want to find your item again, you can just put a product url here, and I have a handy form for adding in my referral code. Or if you want, you can just add it yourself:

&tag=paultarjan-20

Blogging With Jekyll

2010-04-30T00:00:00-07:00

As you may, or may not have noticed, I haven’t been blogging much lately. I think this is due in part to how I was trying to use blogger. I was inserting my own html into the text box while blogging, and it was just a real pain to build a blog post. I had lots of control, but it was too much of a pain (and I’m lazy, so it didn’t happen often).

I’ve also been really admiring clean and simple blog templates whenever I see them, and switching to that in blogger would require me to backport little things I would do in each post.

So, I bit the bullet. I converted all my posts to Markdown. I originally planned on doing a Markdown -> HTML conversion and posting that to blogger, but then I realized that I could just use jekyll to build the whole thing. I lose a lot of the plugins in blogger (that I wasn’t using much anyways) but I gain a lot of simplicity. And now I can write my blog posts in vim and host them as straight .html files! And their source is available! And… and… boy am I excited!

I exported my existing blog (Settings -> Basic -> Export Blog) to an .xml file. I then imported this file using tannerburson’s converter. I lost the command out of my history, but maybe a ruby enthusiast can let me know. Something like:

$ ruby ...stuff... "Jekyll::Blogger.process('feed.xml')"

It made me a _posts directory. I then hand tuned each file to be in markdown instead of html and voila, you now have my new blog. I made jekyl match the permalink structure of my old blog, so no links should break. Let me know if anything goes wrong.

Hopefully I will post more with this new setup. If anyone have any subject requests let me know, and general comments are always welcome (like the format, hate the format, the RSS is broken, etc).

P.S. I wrote this whole post in markdown using vim, and it was MUCH nicer than doing it in blogger’s textarea.

Twitter inside Facebook

2010-04-29T00:00:00-07:00

With the launch of Facebook’s like button, I hoped I could do something that I’ve wanted to do for a while. Create a mirror page of each twitter page inside of facebook. That way, I can just “like” that page, and now I don’t have to have to check facebook AND twitter.

And voila! It actually worked! Try it for yourself.

Twitter 2 Facebook

You can either “like” individual people (like mine) or choose from all your existing followees. Once you click “like” then all twitter posts by that person will show up in your facebook feed. You can unlike it any time (from your profile, or the original page).

If this takes off, then I’ll buy a better domain for it. Any suggestions for the name (that are avilable)?

For the geeks: this uses PHP to proxy the twitter pages, and node.js for the backend. Each proxy page has open graph data embed in it, so that when a person “likes” it, they give my facebook app permission to post to their feed. When you like a page, you ping my backend telling it to start listening to that user’s tweets. My backend then, connects to the twitter streaming API and sets the follow list to all the people who have been “liked”. When a tweet comes in, it goes right back out to the corresponding facebook page using the stream.publish API.

As a byproduct, I built a node.js cache that you might find useful. When a cache key expires it actually purges it, so you don’t end up with tons of memory usage, nor need garbage collection.

Goodbye Yodel, Hello Friend

2010-03-26T00:00:00-07:00

Today is my last day at Yahoo!

I made a video to commemorate the occasion and reminisce over my last 1032 days being a Yahoo.

Co-starting SearchMonkey and being the Chief Technical Monkey was a dream come true, and I couldn’t have asked for a better manager and team. Traveling with HackU was icing on my already delicious (cup)cake.

Cheers to all my Yahoo! friends and I look forward to my new adventure at Facebook.

Rock Band Group Algorithm

2009-10-30T00:00:00-07:00

I often play Rock Band with my friends, but we have a tough time deciding who will play what part and for how long. Here is what we’ve come up with as requirements

Everyone must play every instrument an equal amount of time - To avoid hogging
Every time a person is on an instrument, they must play with a different combination of people - To play on other people’s strengths and avoid other’s weaknesses
Sitting out many songs in a row is minimized

Does anyone else have other constraints they play with?

Given these requirements, here is what we do:

Lay the 4 instruments out and randomly put 4 people on them, and intersperse the remaining people in the gaps between instruments
Cycle everyone through every instrument and every gap (clockwise), until we arrive back at the initial state
Shuffle everyone like this:
1. Group the people into 2 groups. Those on instruments (size 4) and those not (size n-4)
2. Pick 4 people from the non-instruments and randomly put them on instruments
3. If you have less than 4 people on non-instruments, select random people to stay on instruments, but change their instruments randomly
4. Randomly permute the remaining people into the open spots
Play another round and then re-cycle

Notes

Designate one guitar as the Guitar, and one as the Bass. We choose the one with the solo fretboard as the Guitar.
The singer picks the song since it is the hardest to do if you don’t know the song.
If you fail a song, if you are <50% complete, you play again, if you are >50% you move to the next instruments. If you fail again, you add your %s together.
We don’t replay the same song that we already played unless we REALLY need to.

This can work for other games, like Guitar Hero or even non-music games. Just change the number 4 to however many slots you have and run it.

I’m interested in how other people do this. I like fairness, but I also like not having to have a complicated system that requires white boards, a master planner, and a dictator.

And what kind of geek would I be if I didn’t have some code to solve this problem

$ python rockband.py Paul Michelle Martin Melanie Surbhi Ziga Emily
Drums:  Martin
Vocals: Ziga
 Gap:   Melanie
Guitar: Surbhi
 Gap:   Paul
Bass:   Michelle
 Gap:   Emily
Press <Enter> once you finished a round: 

Drums:  Paul
Vocals: Ziga
 Gap:   Surbhi
Guitar: Emily
 Gap:   Michelle
Bass:   Melanie
 Gap:   Martin
Press <Enter> once you finished a round: 

Drums:  Surbhi
Vocals: Martin
 Gap:   Paul
Guitar: Michelle
 Gap:   Melanie
Bass:   Emily
 Gap:   Ziga
Press <Enter> once you finished a round: q

and in rockband.py

import random
import sys

instruments = ["Drums", "Vocals", "Guitar", "Bass"]

people = sys.argv[1:]
if len(people) == 0:
    print "python rockband.py Name1 Name2 ..."
    sys.exit()
if len(people) < len(instruments):
    print "If there is less than %d people, just pick some instruments, "+\
      "rotate, and then shuffle. You don't need me for that"\
      % len(instruments)
    sys.exit()

random.shuffle(people)
num = len(people)

places = [' Gap']*num
instruments_ids = [0, int(num*0.25), int(num*0.5), int(num*0.75)]
for i, ins in zip(instruments_ids, instruments):
    places[i] = ins

while True :
    for l,n in zip(places, people):
        print "%s:\t%s" % (l, n)

    if raw_input("Press <Enter> once you finished a round: ") == "q" :
        break

    print ""
    playing = [people[x] for x in instruments_ids]
    not_playing = list(set(people) - set(playing))

    if len(not_playing) < len(instruments) :
        new_playing = not_playing +\
          random.sample(playing, len(instruments) - len(not_playing))
    else :
        new_playing = random.sample(not_playing, len(instruments))

    new_not_playing = list(set(people) - set(new_playing))
    random.shuffle(new_playing)
    random.shuffle(new_not_playing)

    people = [None]*num
    for i, p in zip(instruments_ids, new_playing) :
        people[i] = p

    count = 0
    for i in xrange(len(people)) :
        if not people[i] :
            people[i] = new_not_playing[count]
            count += 1

This code doesn’t work very well with 9 people as they will always be playing with the same group, but at least their instruments will change. All other sizes should work well.

File Extensions on the Internet

2009-10-20T00:00:00-07:00

I simply wanted to know Which file extensions are used on the internet?

So I wrote a little program (webExtension.py) and a half million calls to Google later, we have some interesting data.

First, the raw data:

Alphabetical : 1 2 3 4 All All (Pruned)
Numerical : 1 2 3 4 All All (Pruned)

html	6 700 000 000
php	5 980 000 000
htm	1 690 000 000
asp	1 510 000 000
aspx	1 380 000 000
jsp	565 000 000
cfm	385 000 000
pdf	298 000 000
do	242 000 000
cgi	199 000 000

Some interesting facts I saw :

There are 1305 unused 3 letter extensions out of the possible 17,576. That is 92.5% are already used for something. (There IS a lot of junk thought, so don’t be TOO alarmed).
There are a lot of common extensions that I have NO idea what they are for. .e? .nhn?
4x more pages are html instead of just htm.
PHP is beating ASP by about 2x.
Many servers serve HTML from image extensions, and jpg > png == gif > svg > jpeg > bmp > tiff
Naming is mostly not biased by first letter. The empty part is 3 letter extensions starting with y.
Only the top 5,000 extensions have more than 1000 pages.

Some caveats

This was done in October 2009, things might change. I’ll rerun it if people leave comments.
I only looked for extensions up to 4 letters. No numbers or funky symbols.
I am assuming the counts on Google’s search results are ACTUALLY correct.

If anyone makes any interesting observations with this data, please let me know and I’ll post it here. Pretty graphs are welcome as well :)

Hadoop Hacking on Yahoo! Ad Data

2009-09-30T00:00:00-07:00

At the CMU Hackday we’re letting students play with an anonymized snapshot of our advertising data. (If you want access, email Jamie, sign something, and we’ll give you a key).

Basically, we have a cluster of EC2 machines running hadoop with the data loaded to play with. So, of course, I wanted to play.

Here is the README about the data

(1) "ydata-ysm-keyphrase-bid-imp-click-v1_0.gz" contains the following fields:

0 day
1 anonymized account_id
2 rank
3 anonymized keyphrase (expressed as list of anonymized keywords)
4 avg bid
5 impressions
6 clicks

Snippet:

1       08bade48-1081-488f-b459-6c75d75312ae    2       2affa525151b6c51 79021a2e2c836c1a 327e089362aac70c fca90e7f73f3c8ef af26d27737af376a    100.0     2.0     0.0
29      08bade48-1081-488f-b459-6c75d75312ae    3       769ed4a87b5010f4 3d4b990abb0867c8 cd74a8342d25d090 ab9f74ae002e80ff af26d27737af376a    100.0     1.0     0.0
29      08bade48-1081-488f-b459-6c75d75312ae    2       769ed4a87b5010f4 3d4b990abb0867c8 cd74a8342d25d090 ab9f74ae002e80ff af26d27737af376a    100.0     1.0     0.0
11      08bade48-1081-488f-b459-6c75d75312ae    1       769ed4a87b5010f4 3d4b990abb0867c8 cd74a8342d25d090 ab9f74ae002e80ff af26d27737af376a    100.0     2.0     0.0

I like python, and hadoop streaming lets me use it for map reducing. You basically write two scripts, map.py and reduce.py which work on stdin, and stdout. Here is me making those files

mkdir money_made_rank
vim money_made_rank/map.py
<code it>
vim money_made_rank/reduce.py
<code it>

map.py:

#!/usr/bin/python
import sys
import random

for line in sys.stdin :
 line = line.strip()
 cols = line.split('\t')
 if len(cols) != 7 :
  continue
 day, account, rank, keyprase, bid, impressions, clicks = cols
 clicks = float(clicks)
 if clicks == 0 :
  continue
 bid = float(bid)
 money = clicks * bid
 print "%s\t%f" % (rank, money)

reduce.py:

#!/usr/bin/python
import sys
from operator import itemgetter

result = {}

for line in sys.stdin :
 line = line.strip()
 key, money = line.split('\t', 1)
 try :
  money = float(money)
  result[key] = result.get(key, 0) + money
 except :
  continue
 
sorted_result = sorted(result.items(), key=itemgetter(0))
for key, money in sorted_result :
 print "%s\t%f" % (key, money)

Then you should test your stuff locally. For that, we left the .gz file and I just ran :

zcat /mnt/data/ydata-ysm-keyphrase-bid-imp-click-v1_0.gz | head -n 1000 | money_made_rank/map.py | sort | money_made_rank/reduce.py

And if it spits out

1       3540.000000
2       14604.489767
3       13516.602689
4       2668.682927
5       2250.000000
6       540.000000
7       540.000000

then you’re doing it right.

If you want to run on one machine then just take out the head -n 1000. That should take about 20 minutes to chew through all the data.

Lets move to hadoop

Once it works in the piping mode, then it is very simple to just do it on the cluster. Don’t change any files, just type :

hadoop jar /usr/lib/hadoop/contrib/streaming/hadoop-0.18.3-14.cloudera.CH0_3-streaming.jar 
  -input /data/ydata/* 
  -output money_made_rank 
  -mapper money_made_rank/map.py 
  -reducer money_made_rank/reduce.py 
  -file money_made_rank/*

This will print out a whole bunch of stuff and at the end you should have a money_made_rank directory. Just print it and bask in the glory:

hadoop fs -cat 'money_made_rank/part-*' | sort -n

and it should print out :

1 14743915410.559452
2 5671857020.978109
3 3580521727.805751
4 1770068342.652887
5 1141008200.372228
6 531839794.947136
7 360624250.037246
8 266172741.734491
9 213458413.893067
10 189563018.302472
...

And then you can put it in a spreadsheet and make a pretty chart. Did you know half of our money comes from the #1 Search Ad?

Converting from MyISAM to InnoDB takes a long time

2009-09-15T00:00:00-07:00

Wow, I didn’t think that with around 80 million rows, MySQL would take 7 hours to convert from MyISAM to InnoDB.

mysql> alter table metaward_achiever ENGINE=INNODB; Query OK, 76756189 rows affected (6 hours 53 min 57.07 sec) Records: 76756189 Duplicates: 0 Warnings: 0

mysql> show create table metaward_achiever; | metaward_achiever | CREATE TABLE metaward_achiever ( id int(11) NOT NULL AUTO_INCREMENT, modified datetime NOT NULL, created datetime NOT NULL, award_id int(11) NOT NULL, alias_id int(11) NOT NULL, count int(11) NOT NULL, PRIMARY KEY (id), KEY metaward_achiever_award_id (award_id), KEY metaward_achiever_alias_id (alias_id) ) ENGINE=InnoDB AUTO_INCREMENT=77166947 DEFAULT CHARSET=utf8 |

Paul Is A Nerd

2009-09-07T00:00:00-07:00

Good thing I named it paulisageek.com, I’m no dork dweeb or nerd!

List of Guilded WoW players

2009-08-26T00:00:00-07:00

I needed a list of as many World of Warcraft players as I could find. Sadly, blizzard wasn’t giving this out on wowarmory and neither were any other sites I could find. Thankfully, I’m good at web scraping.

So, I went through each region, server, guild, and character on wowjustsu and pulled out 1.75 Million characters.

1,019,737 US wowarmory links (gz) (zip)
741,321 EU wowarmory links (gz) (zip)

World of warcraft claims over 10 Million players, wowjutsu claims 4 Million users but I could only find 1.75 Million. Maybe the rest are unguilded or their guilds aren’t listed. Either way, 1.75 Million is good enough for me, hopefully it helps you out.

stdicon

2009-08-18T00:00:00-07:00

From the idea of progrium I did a fun little hack last night.

(mimetype or file extension) -> icon == stdicon.com

In the vein of gravatar’s simple URLs, just add the file extension, or mimetype onto the path, and you will get a good icon representing it. There are more options for choosing your icon set, size, and default, but head over to the root page to find out more.

It’s open source, so fork and fix. And if you have any ideas for good icon sets, please post a comment or email me and I’ll get them into the system.

We won hack day!

2009-08-16T00:00:00-07:00

Friday was my favorite quarterly event at Yahoo, hackday! It’s the internal version of the hacku event that I help with. I want to chronicle my experience since it was my favourite hackday to-date. I’m being overly vague about the project since it might actually get shipped.

This year, I had an idea mulling in my head, talked to a few people, shared it on the internal hacker list and got a ton of great buzz. I started the ball rolling with some editorial work and design. Then our SearchMonkey community manager, Evan Goer lept on the idea, and did a ton more of the editorial work. His enthusiasm was contagious.

When hackday finally came around, Evan was definitely onboard, but it was still a daunting task. He wanted to help in any way he could so I gave him a small coding task in PHP. Lo-and-behold he pulled it off with flying colors. I think we should get huge bonus points for a project manager coding on the project :) He sat next to me at the start of hackday, and went out for dinner, and then CAME BACK to finish it up. What a trooper.

Right as I started to code, one of the SearchMonkey ops guys, Brett Proctor IMed me offering his services. He told me he can barely code and had to leave that night at 9pm, so I gave him some “store and retrieve” webservice to work on. He pulled it off amazingly, and again, we should get more bonus points for an operations guy building the whole backend :)

Prior to hackday, in the email exchange, the SearchMonkey UI guy (notice a trend here?) Micah Alpern suggested it to one of his designers, Kara Mccain. She’s the one that did the SearchMonkey logo, so I was elated to have her onboard. She cooked up some sexy designs for the project and then passed out from the exhaustion that the designs caused :) She stopped by the hack room, and then left before Evan got there, so as far as the rest of the team was concerned, she was “remote” much like Brett.

While Brett was doing his work, he showed the project to Reid Burke who then obviously wanted in. The only non-SearchMonkey guy, but we won’t hold it against him. Reid came to the hack room after Evan left, and hung out with me until the wee hours of the morning. Reid came into the project with a self-contained piece so it was easier to integrate. By the time I left at 5:00 am, Reid’s part still wasn’t working, but I guess my threats to take him off the team worked, because when I came back at noon the next day, Reid’s part magically worked. :)

I’d like to note that there was a handful of people in the hack room that contributed in one way or another to the hack. Matt Claypotch made up a lot of great titles, had some witty banter all night and made us a pretty picture too! Philip Tellis was Mr. Knowledgable sitting in the corner. I would ask a question to the air, Matt would say something snarky, Reid would join in, and then when it simmered down, Philip would quietly answer the question wonderfully. Oh, and Eric Wu was hammering away on the designs for all the little things to make hackday work. Thanks guys, the hack room was great!

We were presenting as #48 out of 92. #46 was my other hack with Yury Lifshits, webnumbr.com. And then #47 was Reid’s other hack. So a big showing all in a row kind of worried me a bit. Right before I was going to present, Eric Wu called a break for pizza. So I went back, got my carb and sugar rush going, and then setup to be first after the break. It was kind of nice placement, since people actually were ready to pay attention. Not to mention I was wearing my wizard hacking hat, which helped the intrigue.

I started off pointing out Jerry Yang was on his 3rd beer, so that boded well for his coercibility. 10 seconds in, I realized I didn’t clear my cookies from all the testing I just did, so nothing was working. I mentioned that noone should pay attention to the man behind the curtain as I quickly ⌘, -> clear cookies. The presentation went well with lots of laughs and clapping. Lots of people in the audience smiled at me and gave me high fives. It was good. Brett couldn’t get the live streaming working, so he tail -f‘ed the apache logs to see what I was showing :)

When the judges came back we were just sitting around after Havi Hoffman gave away some old shirts. I had been to many hack days before this and hadn’t won, so I was already over the emotionally crushing experience of not winning and didn’t care anymore. When Prabhakar Raghavan announced the most innovative entry going to webnumbr, Yury and I were elated! We skipped up and got our awesome bright orange t-shirts with pride. I was just so excited I didn’t pay attention to when Ash Patel gave away the prize for the most fun hack to us! Sadly Brett and Kara couldn’t be there, but Evan and Reid and I bounded up to get our shirts (I didn’t take a second). We got our picture taken as we were hugging and it was just a great experience.

I chatted for a while with everyone and then walked back with Yury and Evan, talking about the future of our stuff. I’m sure on Monday the joy will wear off and the reality of our real jobs will set in, but maybe, just maybe, we can start another small project like SearchMonkey and see where it goes.

So yes, hack day was awesome, and I hope this tradition spreads and thrives.

Online YAML parser

2009-08-12T00:00:00-07:00

So, today I needed to verify some YAML was correct, and eyeball the JSON output. I couldn’t find a good tool online that did what I wanted, so I wrote my own in about an hour.

I give you the Online YAML Parser.

It takes in YAML and outputs JSON using pyYAML. Simple, but useful. Try some examples from the 1.2 spec, or paste in your own, and let me know of any bugs.

SearchMonkey Object Examples

2009-07-29T00:00:00-07:00

So, a few weeks ago we released SearchMonkey Objects that let web developers markup any of our 8 object formats for their pages. These pages included examples and explanations.

To help people understand it a bit more, I marked up a few real-life pages using our SearchMonkey objects.

(games and documents are identical to video).

So, I hope those help a bit. Try running them through the extraction tool. Let me know if you have any issues or have any more examples.

Only 1 word left on google voice, and its mine!

2009-07-16T00:00:00-07:00

I recently got an invitation to google voice and have been trying to come up with a good number. I tried all the first 10 digits of physics constants and every single 10 letter word in the ubuntu /usr/share/dict/words. Out of all of that, there was ONLY 1 that was available.

So I give you my new phone number (which is very easy to memorize, hehe) :

(MEM) ORI-ZING 
(636) 674-9464

The 2 other leading candidates were :

(209) 782-7526 (209) PTA-RJAN

(707) 728-5468 (707) PAU-LIN8 (thanks ultramegaman)

Is it too much to ask for the number (314) 159-2653, (299) 792-4580, or (137) 035-9990? What is a geek to do…

Here is the source of my script:

# This script will search google voice to see if there are any numbers
# available that spell a 10 letter word 
#
# It can easily be customized for 9 letters, or to use your own dictionary, etc.
#
# Author: Paul Tarjan <http://paulisageek.com>
#
# Make sure you put your own cookie here. Search the web if you don't know how to find it.
cookie = "DQAAA<rest of cookie>"

import urllib2
import re

def check(c) :
    global cookie
    r = urllib2.Request("https://www.google.com/voice/setup/search/?q=" + c, headers={"Cookie" : "gv=" + cookie})
    response = urllib2.urlopen(r)
    data = response.read()
    if '"Sign in"' in data:
        print "Bad cookie (probably)"
        return False
    if data.find('"num_matches":"0"') != -1 :
        return False
    return True

def dict(f="/usr/share/dict/words") :
    for word in file(f).read().split() :
        if len(word) == 10 and re.match("^[a-zA-Z0-9]+$", word) :
            if check(word) :
                print word

if "<rest of cookie>" in cookie:
    print "Please look up your google cookie and put it at the top of this file before running"
else:
    dict()

If it can help anyone, here is the list of the first 10 digits of all of the physics constants on that site which are known to 10 digits or greater :

What is "git"

2009-07-15T00:00:00-07:00

I was just compiling git and found this at the top of the README file :

"git" can mean anything, depending on your mood.

 - random three-letter combination that is pronounceable, and not
   actually used by any common UNIX command.  The fact that it is a
   mispronunciation of "get" may or may not be relevant.
 - stupid. contemptible and despicable. simple. Take your pick from the
   dictionary of slang.
 - "global information tracker": you're in a good mood, and it actually
   works for you. Angels sing, and a light suddenly fills the room.
 - "goddamn idiotic truckload of sh*t": when it breaks

Finally, I know how to swear correctly when git isn’t working!

OpenID Avatar

2009-07-03T00:00:00-07:00

gravatar does a wonderful job with email avatars, but they have failed to support OpneIDs. I built a simple interface for associating your OpenID with an email, which is then proxied to gravatar.

So I give you OpenID Avatar.

Simply change your gravatar urls to use http://paulisageek.com/openidavatar/avatar/ instead of http://www.gravatar.com/avatar/ Then you can use md5s of openids in addition to emails. If you want to use them both, just set the email_md5 param as well. See the img in this post for an example, or see the docs.

To setup your openid to email link, simply login and pick which email to associate with. Done.

Suggestions are always welcome.

Easy OpenID Delegation with Yadis

2009-06-29T00:00:00-07:00

Lets say you have a site somewhere that you want to use as your openid. The easiest way to delegate OpenID is to put these in your <head>

<link href='http://www.myopenid.com/server' rel='openid.server'/>
<link href='http://ptarjan.myopenid.com/' rel='openid.delegate'/>

That requires that the URL you are putting them on returns HTML. For me, I have a 302 redirect from http://paulisageek.com to http://blog.paulisageek.com so all of my enpoints are getting my identity as http://blog.paulisageek.com. Not what I wanted.

Enter : Yadis. I created a small Yadis file that says the same thing as those link elements. openid.xml.

<?xml version="1.0" encoding="UTF-8"?>
<xrds:XRDS xmlns:xrds="xri://$xrds" xmlns="xri://$xrd*($v*2.0)" xmlns:openid="http://openid.net/xmlns/1.0">
<XRD>
    <Service priority="10">
        <Type>http://openid.net/signon/1.0</Type>
        <URI>http://www.myopenid.com/server</URI>
        <openid:Delegate>http://ptarjan.myopenid.com/</openid:Delegate>
    </Service>
    <Service priority="20">
        <Type>http://openid.net/signon/1.0</Type>
        <URI>http://draft.blogger.com/openid-server.g</URI>
        <openid:Delegate>http://paulisageek.blogspot.com/</openid:Delegate>
    </Service>
</XRD>
</xrds:XRDS>

And then redirect it if the HTTP Accept header is application/xrds+xml. This is my index.php on paulisageek.com :

<?php
if (strpos($_SERVER['HTTP_ACCEPT'], "application/xrds+xml") !== FALSE) {
    header("Content-Type: application/xrds+xml");
    echo file_get_contents("openid.xml");
    die();
} else {
    header("Location: http://blog.paulisageek.com");
    die();
}
?>

And Voila, it work as an openid endpoint.

Update: You can also use mod_rewrite to get people to your openid.xml file :

    RewriteCond %{HTTP_ACCEPT} application/xrds\+xml
    RewriteRule .* openid.xml [T=application/xrds+xml,L]

You can replace .* with the urls that you want to be allowed openids, but I intentionally have .* so that I can have unlimited openids for myself.

Equivalent OpenIDs

2009-06-28T00:00:00-07:00

I’ve been having a big dilemma whenever I use an openid. Which do I pick from the 8 that could be me? Talking to Joseph Smarr at the Mozilla labs meetup we stumbled upon a solution. rel-me.

I should register on any site with an openid that supports at least 1 rel-me link. Most sites where you can set your “homepage” will add the attribute rel="me" on the link. That basically says that “I am also over there”. If you can only link to one homepage, just make sure that from that homepage you can link out to other things you own on the internet. If you can put multiple rel="me" links, then just list everything you own. I link to my friendfeed.com/ptarjan which has rel="me" links to all my other “stuff”.

Then when I log into any openid provider they should check the social graph to see if the openid I just used is equivalent to any other one. Google made a simple API to query the graph. This has one problem in that they crawl the graph recursivly on outbound links. Meaning they are trying to start at a webpage and find other things from there. That doesn’t stop me from saying that twitter.com/cnn is me. For identity you want to crawl the graph backwards. Similar as to how PageRank only trusts inbound links.

I made a YQL execute file (raw code) that will compute which URIs trust a given URI.

For example, if I log in with

http://paulisageek.com

then the website that I logged in to should make the YQL call

USE 'http://paulisageek.com/yql-tables/socialgraph/socialgraph.trustsme.xml'; SELECT * FROM socialgraph.trustsme WHERE q='http://paulisageek.com'

which returns something like

<query>
    <results>
        <result>
            <trustsme>http://www.paulisageek.com/</trustsme>
            <trustsme>http://www.linkedin.com/pub/dir/paul/tarjan</trustsme>
            <trustsme>http://stackoverflow.com/users/90025/ptarjan</trustsme>
            <trustsme>http://www.mybloglog.com/buzz/members/ptarjan/</trustsme>
            <trustsme>http://digg.com/users/paralax/</trustsme>
            <trustsme>http://www.google.com/profiles/114701835473476527933</trustsme>
            <trustsme>http://www.flickr.com/photos/ptarjan/</trustsme>
            <trustsme>http://www.linkedin.com/in/paultarjan</trustsme>
            <trustsme>http://twitter.com/ptarjan</trustsme>
            <trustsme>http://github.com/ptarjan</trustsme>
            <trustsme>http://www.hulu.com/profiles/ptarjan</trustsme>
            <trustsme>http://stackoverflow.com/users/90025/paul-tarjan</trustsme>
            <trustsme>http://friendfeed.com/ptarjan</trustsme>
            <trustsme>http://www.google.com/profiles/ptarjan</trustsme>
            <trustsme>http://www.flickr.com/photos/78332988@N00/</trustsme>
            <trustsme>sgn://twitter.com/?pk=14757201</trustsme>
        </result>
    </results>
</query>

and then walk down that list starting at the top, looking at all their user accounts to see if any matches. The first match is the account they should be logged into.

Every one of these trust relationships is explicitly put there by the user by a link with rel="me" so it can be trusted just as much as the original openid provider. For example, if there is a link on http://twitter.com/ptarjan with rel="me" to http://paulisageek.com, my twitter account is giving permission to paulisageek to modify any account it owns. If my twitter account were to be hacked to point to another place with rel="me", that is equivalent to the openid login for that URI to be hacked as well. All sites that I am aware of use the same credentials to edit the rel="me" link as for the openid identity login. With this assumption, the trust relationship is valid.

In conclusion, every site that is using openid, should check inbound rel="me" links to the openid URI. Using the socialgraph API I created a helper REST API that can be used to take any URI and find other URIs that trust it.

Blogger Theme with hAtom Microformat

2009-06-27T00:00:00-07:00

I just updated my theme to be Garland but I found that it didn’t include the hAtom microformat. So I tweaked it a little bit, and now I have a beautiful microformat. If anyone else wants to use it, just copy and paste this into the “edit template” part of your blogger layout.

<?xml version="1.0" encoding="UTF-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns='http://www.w3.org/1999/xhtml' xmlns:b='http://www.google.com/2005/gml/b' xmlns:data='http://www.google.com/2005/gml/data' xmlns:expr='http://www.google.com/2005/gml/expr'>
 <head>
   <b:include data='blog' name='all-head-content'/>
   <title><data:blog.pageTitle/></title>
   <b:skin><![CDATA[/*
-----------------------------------------------
Blogger Template Style
Name:     Garland
Designer: Steven Wittens
URL:      http://acko.net/

Modified by: http://www.blogcrowds.com
Support:     http://www.blogcrowds.com
----------------------------------------------- */

a:link, a:visited {
color: #027ac6;
text-decoration: none;
}

a:hover {
color: #0062a0;
text-decoration: underline;
}

a:active, a.active {
color: #5895be;
}

body {
background: #edf5fa;
color: #494949;
font: 12px/170% Verdana, Arial, Helvetica, sans-serif;
margin: 0;
padding: 0;
}

body.sidebar-left  #footer {
margin-left: -210px;
}

/* So we move the #center container over the sidebars to compensate */
body.sidebar-left #center {
margin-left: -210px;
}

/* And add blanks left and right for the sidebars to fill */
body.sidebar-left #squeeze {
margin-left: 210px;
}

body.sidebar-right #center {
margin-right: -210px;
}

body.sidebar-right #footer {
margin-right: -210px;
}

body.sidebar-right #squeeze {
margin-right: 210px;
}

body.sidebars #center {
margin: 0 -210px;
}

body.sidebars #footer {
margin: 0 -210px;
}

body.sidebars #squeeze {
margin: 0 210px;
}

.meta{margin:0 0 30px 0}

.sidebar .widget{margin:0 0 20px 0}

dl {
margin: 0.5em 0 1em 1.5em;
}

dl dd {
margin: 0 0 .5em 1.5em;
}

dl dt {
}

h1 {
font-size: 170%;
}

h1, h2, h3, h4, h5, h6 {
font-family: Helvetica, Arial, sans-serif;
font-weight: normal;
margin: 0;
padding: 0;
}

h2 {
font-size: 160%;
line-height: 130%;
}

h3, .sidebar h2 {
font-size: 140%;
}

h4 {
font-size: 130%;
}

h5 {
font-size: 120%;
}

h6 {
font-size: 110%;
}

hr {
background: #5294c1;
border: none;
height: 1px;
margin: 0;
padding: 0;
}

img, a img {
border: none;
}

input {
color: #494949;
font: 12px/100% Verdana, Arial, Helvetica, sans-serif;
}

p {
margin: 0.6em 0 1.2em;
padding: 0;
}

textarea, select {
color: #494949;
font: 12px/160% Verdana, Arial, Helvetica, sans-serif;
}

.content ul {
margin: 0.5em 0 1em;
padding: 0;
}

.content ul li {
margin: 0.4em 0 0.4em 1.5em;
}

ul, quote, code, fieldset {
margin: .5em 0;
}

/**
* Primary navigation
*/
ul.primary-links {
float: right;
margin: 0;
padding: 0;
position: relative;
z-index: 4;
/* Hide links - no idea how to get rid of them */
       display : none;
}

ul.primary-links li {
background: none;
float: left;
margin: 0;
padding: 0;
}

ul.primary-links li a, ul.primary-links li a:link, ul.primary-links li a:visited {
background: transparent url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/bg-navigation-item.png") no-repeat 50% 0;
color: #fff;
display: block;
margin: 0 1em;
padding: .75em 0 0;
}

ul.primary-links li a:hover, ul.primary-links li a.active {
background: transparent url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/bg-navigation-item-hover.png") no-repeat 50% 0;
color: #fff;
}

.nextprev {
margin-bottom: 60px;
margin-top: 10px;
}

.alignleft {
float: left;
}

.alignright {
float: right;
clear: both;
}

.commentlist {
list-style: none;
margin: 1em 0 3em;
padding: 0;
}

.commentlist .alt {
background: #edf5fa;
}

.commentlist .avatar {
border: 1px dotted #ccc;
float: right;
margin-right: 25px;
padding: 2px;
}

.commentlist cite {
font-weight: bold;
}

.commentlist li {
padding: 1em 2em;
}

.menu {
list-style: none;
margin: 0;
padding: 0;
}

.menu li li, .item-list ul li, li.leaf, .links li, .sidebar .li {
background: transparent url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/menu-leaf.gif") no-repeat 0 .35em;
list-style-image: none;
list-style-type: none;
padding: 0 0 .3em 13px;
}

.post {
margin-bottom: 2em;
overflow: hidden;
}

#header h1, #header h1 a {
color: #fff;
font-size: 1.5em;
font-weight: normal;
text-decoration: none;
text-shadow: #1659ac 0px 1px 3px;
}

#navigation {
background: url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/bg-navigation.png") repeat-x 50% 100%;
height: 1em;
}

#s {
margin-bottom: 5px;
}

/* Layout and images */
#wrapper {
background: url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/body.png") #edf5fa repeat-x 50% 0;
height: 600px;
}

#wrapper #container {
margin: 0 auto;
max-width: 1270px;
padding: 0 20px;
}

/* We must define 100% width to avoid the body being too narrow for near-empty pages */
#wrapper #container #center {
float: left;
width: 100%;
}

/* Now we add the backgrounds for the main content shading */
#wrapper #container #center #squeeze {
background: url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/bg-content.png") #fff repeat-x 50% 0;
}

#wrapper #container #center .right-corner {
background: transparent url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/bg-content-right.png") no-repeat 100% 0;
left: 10px;
position: relative;
}

#wrapper #container #center .right-corner .left-corner {
background: transparent url("http://s61.photobucket.com/albums/h59/c_stavanger/garland/bg-content-left.png") no-repeat 0 0;
left: -10px;
margin-left: -10px;
min-height: 400px;
padding: 60px 25px 5em 35px;
position: relative;
}

#wrapper #container #footer {
clear: both;
color: #898989;
float: none;
margin: 4em 0 -3em;
text-align: center;
}

#wrapper #container #header {
height: 80px;
}

#wrapper #container #header #logo-floater {
position: absolute;
}

#wrapper #container #header h1, #wrapper #container #header h1 a:link, #wrapper #container #header h1 a:visited {
font-weight: bold;
line-height: 120px;
position: relative;
white-space: nowrap;
z-index: 2;
}

/* We ensure the sidebars are still clickable using z-index */
#wrapper #container .sidebar {
float: left;
margin: 60px 0 5em;
width: 208px;
z-index: 2;
}

body > #wrapper #container .sidebar {
position: relative;
}

#sidebar-left li {
margin-right: 10px;
}

#sidebar-right ul {
margin-left: 20px;
}

#sidebar2 {margin:0 0 0 20px;}

#sidebar-right li ul {
margin-left: 0;
}

.sidebar ul {
margin-left: 0;
padding: 0;
}

.content img {
margin-left: 10px;
}

#comment {
width: 400px;
}

]]></b:skin>
 </head>

<body class='sidebars'>
 <div id='navigation'/>

<div id='wrapper'>
<div class='clear-block' id='container'>

   <!-- skip links for text browsers -->
   <span id='skiplinks' style='display:none;'>
     <a href='#main'>skip to main </a> |
     <a href='#sidebar'>skip to sidebar</a>
   </span>

   <div id='header'>
     <b:section class='header' id='header' maxwidgets='1' showaddelement='no'>
<b:widget id='Header1' locked='true' title='paulisageek (Header)' type='Header'>
<b:includable id='title'>
 <b:if cond='data:blog.url == data:blog.homepageUrl'>
   <data:title/>
 <b:else/>
   <a expr:href='data:blog.homepageUrl'><data:title/></a>
 </b:if>
</b:includable>
<b:includable id='description'>
 <div class='descriptionwrapper'>
   <p class='description'><span><data:description/></span></p>
 </div>
</b:includable>
<b:includable id='main'>
 <div id='logo-floater'>
   <h1>
      <b:if cond='data:blog.url == data:blog.homepageUrl'>
       <a href='#'><data:title/></a>
     <b:else/>
       <a expr:href='data:blog.homepageUrl'><data:title/></a>
     </b:if>
   </h1>
 </div>
 <ul class='links primary-links'>
   <li><a expr:href='data:blog.homepageUrl'><data:title/></a></li>
 </ul>
</b:includable>
</b:widget>
</b:section>
   </div>

   <div class='sidebar' id='sidebar-left'><b:section class='side' id='sidebar1' preferred='yes'>
<b:widget id='Feed1' locked='false' title='FriendFeed - ptarjan' type='Feed'>
<b:includable id='main'>
   <h2><data:title/></h2>
   <div class='widget-content' expr:id='data:widget.instanceId + "_feedItemListDisplay"'>
     <span style='filter: alpha(25); opacity: 0.25;'>
       <a expr:href='data:feedUrl'><data:loadingMsg/></a>
     </span>
   </div>
   <b:include name='quickedit'/>
 </b:includable>
</b:widget>
<b:widget id='BlogArchive1' locked='false' title='Blog Archive' type='BlogArchive'>
<b:includable id='main'>
 <b:if cond='data:title'>
   <h2><data:title/></h2>
 </b:if>
 <div class='widget-content'>
 <div id='ArchiveList'>
 <div expr:id='data:widget.instanceId + "_ArchiveList"'>
   <b:if cond='data:style == "HIERARCHY"'>
    <b:include data='data' name='interval'/>
   </b:if>
   <b:if cond='data:style == "FLAT"'>
     <b:include data='data' name='flat'/>
   </b:if>
   <b:if cond='data:style == "MENU"'>
     <b:include data='data' name='menu'/>
   </b:if>
 </div>
 </div>
 <b:include name='quickedit'/>
 </div>
</b:includable>
<b:includable id='flat' var='data'>
 <ul>
   <b:loop values='data:data' var='i'>
     <li class='archivedate'>
       <a expr:href='data:i.url'><data:i.name/></a> (<data:i.post-count/>)
     </li>
   </b:loop>
 </ul>
</b:includable>
<b:includable id='menu' var='data'>
 <select expr:id='data:widget.instanceId + "_ArchiveMenu"'>
   <option value=''><data:title/></option>
   <b:loop values='data:data' var='i'>
     <option expr:value='data:i.url'><data:i.name/> (<data:i.post-count/>)</option>
   </b:loop>
 </select>
</b:includable>
<b:includable id='interval' var='intervalData'>
 <b:loop values='data:intervalData' var='i'>
     <ul>
       <li expr:class='"archivedate " + data:i.expclass'>
         <b:include data='i' name='toggle'/>
         <a class='post-count-link' expr:href='data:i.url'><data:i.name/></a>
           <span class='post-count' dir='ltr'>(<data:i.post-count/>)</span>
         <b:if cond='data:i.data'>
           <b:include data='i.data' name='interval'/>
         </b:if>
         <b:if cond='data:i.posts'>
           <b:include data='i.posts' name='posts'/>
         </b:if>
       </li>
     </ul>
 </b:loop>
</b:includable>
<b:includable id='toggle' var='interval'>
 <b:if cond='data:interval.toggleId'>
 <b:if cond='data:interval.expclass == "expanded"'>
   <a class='toggle' expr:href='data:widget.actionUrl + "&amp;action=toggle" +       "&amp;dir=close&amp;toggle=" + data:interval.toggleId +       "&amp;toggleopen=" + data:toggleopen'>
       <span class='zippy toggle-open'>▼ </span>
   </a>
 <b:else/>
   <a class='toggle' expr:href='data:widget.actionUrl + "&amp;action=toggle" +         "&amp;dir=open&amp;toggle=" + data:interval.toggleId +         "&amp;toggleopen=" + data:toggleopen'>
         <span class='zippy'>
           <b:if cond='data:blog.languageDirection == "rtl"'>
             ◄
           <b:else/>
             ►
           </b:if>
         </span>
   </a>
 </b:if>
</b:if>
</b:includable>
<b:includable id='posts' var='posts'>
 <ul class='posts'>
   <b:loop values='data:posts' var='i'>
     <li><a expr:href='data:i.url'><data:i.title/></a></li>
   </b:loop>
 </ul>
</b:includable>
</b:widget>
<b:widget id='AdSense1' locked='false' title='' type='AdSense'>
<b:includable id='main'>
 <div class='widget-content'>
   <data:adCode/>
 </div>
</b:includable>
</b:widget>
</b:section></div>

     <div id='crosscol-wrapper' style='text-align:center'>
       <b:section class='crosscol' id='crosscol' showaddelement='no'/>
     </div>

     <div id='center'><div id='squeeze'><div class='right-corner'><div class='left-corner'>
     <!-- begin content -->
     <div class='node'>
       <b:section class='main' id='main' showaddelement='no'>
<b:widget id='Blog1' locked='true' title='Blog Posts' type='Blog'>
<b:includable id='nextprev'>
 <div class='blog-pager hentry' id='blog-pager'>
   <b:if cond='data:newerPageUrl'>
     <span id='blog-pager-newer-link'>
     <a class='blog-pager-newer-link' expr:href='data:newerPageUrl' expr:id='data:widget.instanceId + "_blog-pager-newer-link"' expr:title='data:newerPageTitle'><data:newerPageTitle/></a>
     </span>
   </b:if>

   <b:if cond='data:olderPageUrl'>
     <span id='blog-pager-older-link'>
     <a class='blog-pager-older-link' expr:href='data:olderPageUrl' expr:id='data:widget.instanceId + "_blog-pager-older-link"' expr:title='data:olderPageTitle'><data:olderPageTitle/></a>
     </span>
   </b:if>

   <b:if cond='data:blog.homepageUrl != data:blog.url'>
     <a class='home-link' expr:href='data:blog.homepageUrl'><data:homeMsg/></a>
     <b:else/>
     <b:if cond='data:newerPageUrl'>
       <a class='home-link' expr:href='data:blog.homepageUrl'><data:homeMsg/></a>
     </b:if>
   </b:if>

 </div>
 <div class='clear'/>
</b:includable>
<b:includable id='backlinks' var='post'>
 <a name='links'/><h4><data:post.backlinksLabel/></h4>
 <b:if cond='data:post.numBacklinks != 0'>
   <dl class='comments-block' id='comments-block'>
     <b:loop values='data:post.backlinks' var='backlink'>
       <div class='collapsed-backlink backlink-control'>
         <dt class='comment-title'>
           <span class='backlink-toggle-zippy'> </span>
           <a expr:href='data:backlink.url' rel='nofollow'><data:backlink.title/></a>
           <b:include data='backlink' name='backlinkDeleteIcon'/>
         </dt>
         <dd class='comment-body collapseable'>
           <data:backlink.snippet/>
         </dd>
         <dd class='comment-footer collapseable'>
           <span class='comment-author'><data:post.authorLabel/> <data:backlink.author/></span>
           <span class='comment-timestamp'><data:post.timestampLabel/> <data:backlink.timestamp/></span>
         </dd>
       </div>
     </b:loop>
   </dl>
 </b:if>
 <p class='comment-footer'>
   <a class='comment-link' expr:href='data:post.createLinkUrl' expr:id='data:widget.instanceId + "_backlinks-create-link"' target='_blank'><data:post.createLinkLabel/></a>
 </p>
</b:includable>
<b:includable id='post' var='post'>
  <div class="hentry post">
   <a expr:name='data:post.id'/>
   <b:if cond='data:post.title'>
     <h2 class="entry-title post-title">
    <b:if cond='data:post.link'>
      <a expr:href='data:post.link' rel="bookmark"><data:post.title/></a>
    <b:else/>
       <b:if cond='data:post.url'>
         <a expr:href='data:post.url' rel="bookmark"><data:post.title/></a>
       <b:else/>
         <data:post.title/>
       </b:if>
    </b:if>
     </h2>
   </b:if>
  
   <span class='submitted'>
     Posted On <data:post.dateHeader/>
     <b:if cond='data:top.showTimestamp'>
       at
       <data:top.timestampLabel/>
       <b:if cond='data:post.url'>
         <a class='timestamp-link' expr:href='data:post.url' rel='bookmark' title='permanent link'><abbr class='published' expr:title='data:post.timestampISO8601'><data:post.timestamp/></abbr></a>
       </b:if>
     </b:if> by <span class="vcard author fn"><data:post.author/></span>
    
       <!-- email post links -->
       <b:if cond='data:post.emailPostUrl'>
         <span class='item-action'>
         <a expr:href='data:post.emailPostUrl' expr:title='data:top.emailPostMsg'>
           <span class='email-post-icon'> </span>
         </a>
         </span>
       </b:if>

       <!-- quickedit pencil -->
       <b:include data='post' name='postQuickEdit'/>
    
   </span>

   <div class='content entry-content'>
     
<data:post.body/></p>
     <div style='clear: both;'/> <!-- clear for photos floats -->
   </div>
  
   <div class='meta'>
       Posted in  <b:if cond='data:post.labels'>
         <data:postLabelsLabel/>
         <b:loop values='data:post.labels' var='label'>
           <a expr:href='data:label.url' rel='tag'><data:label.name/></a><b:if cond='data:label.isLast != "true"'>,</b:if>
         </b:loop>
       </b:if> |

       <b:if cond='data:blog.pageType != "item"'>

         <b:if cond='data:post.allowComments'>
           <a class='comment-link' expr:href='data:post.addCommentUrl' expr:onclick='data:post.addCommentOnclick'><b:if cond='data:post.numComments == 1'>1 <data:top.commentLabel/><b:else/><data:post.numComments/> <data:top.commentLabelPlural/></b:if></a>
         </b:if>
       </b:if>
  

      <!-- backlinks -->
        <b:if cond='data:blog.pageType != "item"'>
          <b:if cond='data:post.showBacklinks'>
            <a class='comment-link' expr:href='data:post.url + "#links"'><data:top.backlinkLabel/></a>
          </b:if>
        </b:if>
   </div>
  </div>
</b:includable>
<b:includable id='commentDeleteIcon' var='comment'>
 <span expr:class='"item-control " + data:comment.adminClass'>
   <a expr:href='data:comment.deleteUrl' expr:title='data:top.deleteCommentMsg'>
     <span class='delete-comment-icon'> </span>
   </a>
 </span>
</b:includable>
<b:includable id='status-message'>
 <b:if cond='data:navMessage'>
 <div class='status-msg-wrap'>
   <div class='status-msg-body'>
     <data:navMessage/>
   </div>
   <div class='status-msg-border'>
     <div class='status-msg-bg'>
       <div class='status-msg-hidden'><data:navMessage/></div>
     </div>
   </div>
 </div>
 <div style='clear: both;'/>
 </b:if>
</b:includable>
<b:includable id='feedLinks'>
 <b:if cond='data:blog.pageType != "item"'> <!-- Blog feed links -->
   <b:if cond='data:feedLinks'>
     <div class='blog-feeds'>
       <b:include data='feedLinks' name='feedLinksBody'/>
     </div>
   </b:if>

   <b:else/> <!--Post feed links -->
   <div class='post-feeds'>
     <b:loop values='data:posts' var='post'>
       <b:if cond='data:post.allowComments'>
         <b:if cond='data:post.feedLinks'>
           <b:include data='post.feedLinks' name='feedLinksBody'/>
         </b:if>
       </b:if>
     </b:loop>
   </div>
 </b:if>
</b:includable>
<b:includable id='comment-form' var='post'>
 <div class='comment-form'>
   <a name='comment-form'/>
   <h4 id='comment-post-message'><data:postCommentMsg/></h4>
   
<data:blogCommentMessage/></p>
   <data:blogTeamBlogMessage/>
   <a expr:href='data:post.commentFormIframeSrc' id='comment-editor-src'/>
   <iframe allowtransparency='true' class='blogger-iframe-colorize blogger-comment-from-post' frameborder='0' height='275' id='comment-editor' name='comment-editor' scrolling='no' src='' width='100%'/>
   <data:post.friendConnectJs/>
   <data:post.cmtfpIframe/>
   <script type='text/javascript'>
     BLOG_CMT_createIframe('<data:post.appRpcRelayPath/>', '<data:post.communityId/>');
   </script>
 </div>
</b:includable>
<b:includable id='backlinkDeleteIcon' var='backlink'>
 <span expr:class='"item-control " + data:backlink.adminClass'>
   <a expr:href='data:backlink.deleteUrl' expr:title='data:top.deleteBacklinkMsg'>
     <span class='delete-comment-icon'> </span>
   </a>
 </span>
</b:includable>
<b:includable id='feedLinksBody' var='links'>
 <div class='feed-links'>
 <data:feedLinksMsg/>
 <b:loop values='data:links' var='f'>
    <a class='feed-link' expr:href='data:f.url' expr:type='data:f.mimeType' target='_blank'><data:f.name/> (<data:f.feedType/>)</a>
 </b:loop>
 </div>
</b:includable>
<b:includable id='postQuickEdit' var='post'>
 <b:if cond='data:post.editUrl'>
   <span expr:class='"item-control " + data:post.adminClass'>
     <a expr:href='data:post.editUrl' expr:title='data:top.editPostMsg'>
       <span class='quick-edit-icon'> </span>
     </a>
   </span>
 </b:if>
</b:includable>
<b:includable id='comments' var='post'>
   <a name='comments'/>
   <b:if cond='data:post.allowComments'>
     <h3 id='comments'>
       <b:if cond='data:post.numComments == 1'>
         1 <data:commentLabel/>:
       <b:else/>
         <data:post.numComments/> <data:commentLabelPlural/>:
       </b:if>
     </h3>

     <ol class='commentlist'>
       <b:loop values='data:post.comments' var='comment'>
         <li>
           <a expr:name='"comment-" + data:comment.id'/>
           <cite>
           <b:if cond='data:comment.authorUrl'>
             <a expr:href='data:comment.authorUrl' rel='nofollow'><data:comment.author/></a>
           <b:else/>
             <data:comment.author/>
           </b:if>
           </cite> Says:
           <small class='commentmetadata'><a expr:href='"#comment-" + data:comment.id' title='comment permalink'>
               <data:comment.timestamp/>
             </a>
             <b:include data='comment' name='commentDeleteIcon'/>
           </small>
             <b:if cond='data:comment.isDeleted'>
             <span class='deleted-comment'><data:comment.body/></span>
           <b:else/>
             
<data:comment.body/></p>
           </b:if>
         </li>
       </b:loop>
     </ol>

     <p class='comment-footer'>
       <a expr:href='data:post.addCommentUrl' expr:onclick='data:post.addCommentOnclick'><data:postCommentMsg/></a>
     </p>
   </b:if>

   <div id='backlinks-container'>
   <div expr:id='data:widget.instanceId + "_backlinks-container"'>
      <b:if cond='data:post.showBacklinks'>
        <b:include data='post' name='backlinks'/>
      </b:if>
   </div>
   </div>
</b:includable>
<b:includable id='main' var='top'>
 <!-- posts -->
 <div class='post hfeed'>

   <b:include data='top' name='status-message'/>

   <b:loop values='data:posts' var='post'>
     <b:include data='post' name='post'/>
     <b:if cond='data:blog.pageType == "item"'>
       <b:include data='post' name='comments'/>
     </b:if>
   </b:loop>
 </div>

 <!-- navigation -->
 <b:include name='nextprev'/>

 <!-- feed links -->
 <b:include name='feedLinks'/>
</b:includable>
</b:widget>
</b:section></div><!-- end content --><!-- spacer for skins that want sidebar and main to be the same height--><div class='clear'> </div><div id='footer-wrapper'><b:section class='footer' id='footer'/></div></div></div></div></div> <!-- /.left-corner, /.right-corner, /#squeeze, /#center --><div class='sidebar' id='sidebar-right'><b:section class='side' id='sidebar2' preferred='yes'>
<b:widget id='Profile1' locked='false' title='About Me' type='Profile'>
<b:includable id='main'>
   <b:if cond='data:title != ""'>
     <h2><data:title/></h2>
   </b:if>
   <div class='widget-content'>
   <b:if cond='data:team == "true"'> <!-- team blog profile -->
     <ul>
       <b:loop values='data:authors' var='i'>
         <li><a expr:href='data:i.userUrl'><data:i.display-name/></a></li>
       </b:loop>
     </ul>

     <b:else/> <!-- normal blog profile -->

     <b:if cond='data:photo.url != ""'>
       <a expr:href='data:userUrl'><img class='profile-img' expr:alt='data:photo.alt' expr:height='data:photo.height' expr:src='data:photo.url' expr:width='data:photo.width'/></a>
     </b:if>

     <dl class='profile-datablock'>
       <dt class='profile-data'><data:displayname/></dt>

       <b:if cond='data:showlocation == "true"'>
         <dd class='profile-data'><data:location/></dd>
       </b:if>

       <b:if cond='data:aboutme != ""'><dd class='profile-textblock'><data:aboutme/></dd></b:if>
     </dl>
     <a class='profile-link' expr:href='data:userUrl'><data:viewProfileMsg/></a>
   </b:if>

    <b:include name='quickedit'/>
   </div>
 </b:includable>
</b:widget>
<b:widget id='HTML1' locked='false' title='My Other Internet Stuff' type='HTML'>
<b:includable id='main'>
 <!-- only display title if it's non-empty -->
 <b:if cond='data:title != ""'>
   <h2 class='title'><data:title/></h2>
 </b:if>
 <div class='widget-content'>
   <data:content/>
 </div>

 <b:include name='quickedit'/>
</b:includable>
</b:widget>
<b:widget id='Followers1' locked='false' title='Followers' type='Followers'>
<b:includable id='main'>
 <b:if cond='data:title != ""'>
   <b:if cond='data:codeSnippet != ""'>
     <h2 class='title'><data:title/></h2>
   <b:else/>
     <b:if cond='data:totalFollowerCount != ""'>
       <h2 class='title'><data:title/> (<data:totalFollowerCount/>)</h2>
     </b:if>
   </b:if> 
 </b:if>
 <div class='widget-content'>
   <div expr:id='data:widget.instanceId + "-wrapper"'>
     <b:if cond='data:codeSnippet != ""'>
       <div style='margin-right:2px;'>
         <data:codeSnippet/>
       </div>
     <b:else/>
       <b:if cond='data:totalFollowerCount == ""'>
         <span class='item-control following-not-admin'>
           <b><data:failureSnippet/></b>
         </span>
         <span class='item-control blog-admin'>
           <b><data:adminFailureSnippet/></b>
         </span>
       <b:else/>
         <b:if cond='data:followingLinkPresent'>
           <div class='follow-this profile-link item-control following-follow-this'>
             <a expr:href='"javascript:_FollowersView._openPopup(\"" + data:followUri + "\");"'>
               <data:followThisMessage/>
             </a>
           </div>
           <div class='follow-this profile-link item-control following-stop-following-this'>
             <a expr:href='"javascript:_FollowersView._openPopup(\"" + data:followUri + "\");"'>
               <data:stopFollowingMessage/>
             </a>
           </div>
         </b:if>

         <div class='followers-grid'>
           <b:if cond='data:totalFollowerCount == 0'>
             <div class='profile-link item-control following-follow-this'>
               <data:emptyFollowersMessage/>
             </div>
           </b:if>
           <!--
           Relies on the js written out in navbar.gxp
           -->
           <b:loop values='data:followers' var='follower'>
             <div class='follower'>
               <a expr:href='data:follower.profileUrl' expr:title='data:follower.displayName' rel='nofollow'>
               <img class='follower-img' expr:alt='data:follower.displayName' expr:height='data:follower.imageHeight' expr:onerror='"this.onerror=null;this.src=\"" + data:anonFollowerImageUrl + "\";"' expr:onload='"setAttributeOnload(this, \"src\", \"" + data:follower.imageUrl + "\")"' expr:width='data:follower.imageWidth' src='http://img1.blogblog.com/img/blank.gif'/>
               </a>
             </div>
           </b:loop>
           <div class='clear'/>
         </div>

         <div class='followers-canvas profile-link'>
           <data:followersFooterMessage/>
           <span class='item-control following-not-admin'>
             <a expr:href='data:followersUri'>
               <data:viewAllMessage/>
             </a>
           </span>
           <span class='item-control blog-admin'>
             <a expr:href='data:manageFollowersUri'>
               <data:manageFollowersMessage/>
             </a>
           </span>
         </div>
         </b:if>
     </b:if>
   </div>
   <b:include name='quickedit'/>
 </div>
</b:includable>
</b:widget>
</b:section><p style='margin:0 0 0 20px'/></div>
  
   </div> <!-- /container -->
   </div>

</body>
</html>
</code></pre></div>

<script>
function fnSelect(objId)
{
   fnDeSelect();
   if (document.selection) 
   {
      var range = document.body.createTextRange();
      range.moveToElementText(document.getElementById(objId));
      range.select();
   }
   else if (window.getSelection) 
   {
      var range = document.createRange();
      range.selectNode(document.getElementById(objId));
      window.getSelection().addRange(range);
   }
}
function fnDeSelect() 
{
   if (document.selection)
             document.selection.empty();
   else if (window.getSelection)
              window.getSelection().removeAllRanges();
} 

var id = "garland_template";
document.getElementById(id).onclick = function() {
  fnSelect(id);
}
</script>
<!-- Script by hscripts.com -->

Debugging django

2009-06-20T00:00:00-07:00

Simply put

import pdb; pdb.set_trace()

into a file named debug.py somewhere in your path. Then anywhere in your django app you just put

import debug

and when you run that webpage it will seem to take a really long time. Just switch to the console and you will have a python debugger open. I usually just type normal python commands print variable, dir(object) but it also support commands like gdb. c for continue running, n for next line, etc…

Thanks to debugging django for this tip.

Incomplete Format

2009-06-01T00:00:00-07:00

Anyone who is getting ValueError: incomplete format exceptions when doing a string format in python should understand that the part inside the () is just an array lookup. You still need a format type after it.

I have %(count) %(adj) %(noun)

doesn’t work since you didn’t say they were integers and strings. You need to do

I have %(count)i %(adj)s %(noun)s

which is the same as

I have %i %s %s

with position indexed values.

Password Management

2009-05-24T00:00:00-07:00

For the past 10 years I’ve been having a headache managing passwords for websites. I had about 7 levels of passwords and I could never decide which one to use on a website. And even once I decided, I would usually type them all trying to log in. If they logged my password attempts now they can go edit my Facebook profile, watch my Hulu shows, and buy stuff with my PayPal account (true story).

So, I’ve decided to generate a new random password per website and put all my usernames and passwords into an encrypted file on my hard drive. My process was

# gpg passwords.gpg
# vim passwords
# gpg -c passwords
# rm passwords

This works fine, but sometimes I forget the last 2 steps and have a fully decrypted file sitting there… yeah… and the gpg -c would ask for my password 2 times in addition to the first gpg, so I end up typing it 3 times in total.

Enter gpg-encrypt. It’s just a simple shell script that reads in an encrypted file, edits it, then saves it back in an encrypted format.

Every time I have to register for a new website I:

Get a good random password
Run ./gpg-encrypt
Add the new information. I keep my file in YAML format incase I want to read it with a program later. Example:

delicious.com : {
    username : ptarjan,
    password : WqgyzN9bJ6Pm
}

Leave the editor and it goes back to being encrypted.

If I have to log in, I just do the same steps as above, but don’t add anything. I can search using my editor for the domain of the site I’m looking for.

I backup my passwords.gpg file every time I add a new password, to another computer incase my linux box dies.

So, I hope this new password management works better for me than the last system. Anyone have any better solutions to this painful problem? And don’t say openid, until 80% of the sites I visit support it :)

#!/bin/sh
filename="passwords.gpg"
if [ $EDITOR ]
then
    editor=$EDITOR
elif [ -f /usr/bin/editor ]
then
    editor=/usr/bin/editor
else
    editor=vi
fi

if [ $# = 0 ] 
then
    echo "No filename specified. Using default $filename"
elif [ $# > 1 ]
then
    echo "$0 [filename.gpg]"
    exit 2
fi

if [ -f `which shred` ]
then
    rm=shred
else
    rm=rm
fi

tmp=`mktemp` || exit 1

# don't show typing
stty -echo
read -p "Password: " passw; echo
stty echo

if [ ! -f $filename ]
then 
    echo "$filename doesn't exist. Starting from empty file."
elif [ ! -w $filename ]
then
    echo "$filename isn't writable."
    exit 1
else
    # decrypt into the tmp file
    echo "$passw" | gpg -q -d --passphrase-fd 0 $filename > $tmp
    if [ $? != 0 ]; then
        # if gpg didn't work, exit
        $rm $tmp
        exit $?;
    fi
fi

$editor $tmp
echo "$passw" | gpg -q -c --passphrase-fd 0 --output $filename $tmp
$rm $tmp
if [ $? != 0 ]; then
    # if gpg didn't work, exit
    exit $?;
fi

St. Maarten/Martin

2009-05-10T00:00:00-07:00

Trains, Trains, Trains, Trains, Trains, Trains, Planes, Planes, and an Automobile

We left the day before my birthday after work. To get to the airport, we took :

VTA from San Jose to Mountain view
Express Caltrain to somewhere near the airport
Normal Caltrain about 2 stops to Millbrae
BART to some transfer station
Another BART to the Airport
SFO train to the terminal
Airplane to Charlette, NC
Airplane to Phillipsburg, St Martin
Rent-a-car to our hotel

I ended up sleeping in a great rocking chair at the airport while charging my Nintendo DS.

A red-eye is probably not the best way to spend a birthday, but it did the job and we were in paradise in time to enjoy a lobster dinner for my birthday.

A Topless Beach - Europe definitely got it right

The second day we swam a bit in our “∞ pool”

and then took off to Orient Beach. I was briefed by Michelle that part of it would be a nude beech. She forgot to inform to me that it was 98% old fat men, and the other 2% older fat women. Lest I say that I was disappointed and chose to not disrobe in these circumstances.

In dismay we trekked to the clothed part of the beach to discover that it was “Europe” rules for women going topless, and indeed the eye candy was much more enjoyable. Michelle told me that etiquette dictated for me to not take any pictures, so you guys just get to imagine.

Up Close and Personal with Airplanes

Heading to Maho Beach.

We went to Sunset Bar, to have a nice drink and then

we watched the planes fly a few meters over our heads. Very cool :

We got back just in time for the “welcome party” which was an open bar. I was drinking “Wipe Me Out”s and after 4, I was VERY much wiped out. We stumbled back to our place and crashed hard for the night.

French Onion Soup + Escargot

Switching to the French side of the island, was drove to Margiot. ALL the road signs were in French, all the people spoke French, all the food was French. It was GREAT! The Dutch side had converted to an American style, but the French had really kept their culture. And it was magnifique! Sadly the rain followed us around the island, but it didn’t stop us from having a great time.

We had some AMAZING authentic French food on the beach and I got to practice quite a bit of my high school French. It was amusing to listen to all the people around us speaking fluent Parisian French and for me to try and join in :).

After our fill we headed to “Happy Bay.” It started to rain a bit as we started to walk up, but we looked at each other and decided to go for it. It was a 10 minute hike up and ended in a beautiful view. Turning around, it IMMEDIATELY started to pour on us forcing us to run down the hill and jump into the car. Praying that it didn’t leak we sat and waited it out enough for the threadbare tires to grip us enough to get home.

Surf and Jerk

Addicted to the great food, we headed to the beach for a french lunch, and guavaberry coladas. Then to Grand Case (the Caribbean’s gourmet capital) to find … what else, crepes! After putting on our extra 5 pounds we went to Pineapple Pete’s for a great meal.

We shared a their famous lobster thermadore and a jerk chicken sandwich (hence the corny title for this day).

I’m on a Boat

We headed to the French side again this day to take a boat out to an island just off the Orient Beach. Before we stopped off at a truck to have lunch. They had a ton of authentic French food, (evan a Croque-monsieur that we had at our cooking class), but we opted to not ruin our french experience from the back of a van and got some hamburgers.

It was gorgeous out there and we took a plethora of pictures.

They kept their restaurant lobsters out in the ocean and when someone ordered one they would pull in the cage and lift them out.

In Search of Crepes

For our last day we went back to Orient beach and played a little football in the waves.

Of course we had to find an authentic French crepe to end our great vacation.

It was a wonderful birthday trip. If you want to see any other pictures, feel free to check out the full set of pictures.

My Performance Review

2009-03-05T00:00:00-08:00

So this year I’ve got my focal review in the form of a Haiku:

Paul. Smart kid. Search monkey.
Cupcakes, cookies, good girlfriend.
When is next big win?

SearchMonkey Video

2009-02-19T00:00:00-08:00

Hot on the heals of the SearchMonkey song comes the SearchMonkey video. Yahoo! video link.

Lyrics:

SearchMonkey get information
SearchMonkey help you out
SearchMonkey store structured data
That you can't live without.

You say old searching very diligent
But its output stink
Result not functional or elegant
What do SearchMonkey think?

SearchMonkey think maybe developer want to write god damn result page himself
SearchMonkey just say it out loud
SearchMonkey is crazy and proud

SearchMonkey like Yahoo!
SearchMonkey like semantic knowledge
SearchMonkey very simple app
That anyone can use:
Even Monkey like you

SearchMonkey Song

2009-01-30T00:00:00-08:00

I wrote a little song for my product at work, SearchMonkey (or Search Monkey if you want to misspell it).

Enjoy :) And please suggest more verses or make a video with it :)

SearchMonkey song

Lyrics:

SearchMonkey get information
SearchMonkey help you out
SearchMonkey store structured data
That you can't live without.
 
You say old searching very diligent
But its output stink
Result not functional or elegant
What do SearchMonkey think?
 
SearchMonkey think maybe developer want to write god damn result page himself
SearchMonkey just say it out loud
SearchMonkey is crazy and proud
 
SearchMonkey like Yahoo!
SearchMonkey like semantic knowledge
SearchMonkey very simple app
That anyone can use:
Even Monkey like you

Note: This is NOT endorsed by Yahoo! or anything like that. It is just me having a little fun :)