The stats aren’t totally broken, just flaky.

I took a look at the new stat system. It’s going to be great, far more detailed than before. That said it froze my browser (Mozilla) after a few minutes. Its ok. I wrote a few custom scripts to keep track of users on my own. This might actually be even more useful later on.

Destruction leads to a very rough road, but it also breeds creation
-RHCP

I was poking around mail() functions for the researcher authentication system when I discovered reCaptcha. It’s run by a lab at Carnegie Melon. In a nutshell it reads books and when it hits a word it can’t understand, it sends in it as a CAPTCHA [those forms that determine if you're human].

About 60 million CAPTCHAs [aprogram that can tell whether a user is human or robot] aresolved by humans around the world every day. In each case, roughly ten seconds of human time are being spent. Individually, that’s not a lot of time, but in aggregate these little puzzles consume more than 150,000 hours of work each day. What if we could make positive use of this human effort? reCAPTCHA does exactly that by channeling the effort spent solving CAPTCHAs online into “reading” books.

reCAPTCHA improves the process of digitizing books by sending words that cannot be read by computers to the Web in the form of CAPTCHAs for humans to decipher. More specifically, each word that cannot be read correctly by OCR is placed on an image and used as a CAPTCHA. This is possible because most OCR programs alert you when a word cannot be read correctly.

That’s a win-win for me.

I don’t understand the rational behind the new Facebook Apps. Users no longer ‘Add’ or ‘Remove’ apps. Instead a user is asked to ‘authorize’ the app, allowing it access to profile info etc.

This is a real pain in the ass.

Before we used a script (post-add) which would add the user to the database, create an account, and all the other housekeeping. That’s gone. Now I have to add users to the database at a few strategic locations instead. Not a hard to do, but it muddles up things. To boot, there’s no more ‘post-remove’ link. This means I have no idea when a user removes the application. My DB swells, my metrics are unreliable, and a lot of issues are created. Again, frustrating.

The facebook app settings page is a mess of old style config and new style config. Annoying.

But easily the worst move on facebook’s move was breaking stats. How could they release this new style without stats? Word on the street is that they are ‘working on it.’ But really that’s unacceptable.

I’m sure this will be sorted out soon… But we really need some documentation / explanation of the new rational. It’s hard to launch an app when you don’t know what ‘version’ of facebook your users are using. This is incredibly bad timing.

kampylelogo

Kampyle Closes the Feedback Loop, I can’t wait to try it out. Iterate often and quickly, that’s the name of the game. This service lets developers put a customized feedback form on their site, complete with a pretty robust set of management tools. You can even personally email the end-user back.

“Thanks, I didn’t notice that. I’ll fix it tonight”.  Will definitely help foster a better user community.

Using Craigslist as an economic bellwether is a pretty cool idea.

craigslist garage sales

There were a total of 129,653 garage sales posted to Craigslist nationwide in May of 2007. In May 2008 there were 252,561

Bad news for the Fed, good news for StudyBob. I’d imagine a lot of those people having garage sales just might be interested in making some more money on the side…

Google Webmaster Tools, very helpful tools. See your website from the eyes of a google crawler.

Kind of a shocker to me. Seems this blog gets far more attention than any of StudyBob proper. Check this oute:
What Googlebot sees

Ouch. Looks like I need to do some tweaking, badly!

Well the new facebook style got rolled out today (or maybe it’s beta?). It’s mostly a good thing for StudyBob.

The good:

• UI needed to be… refined… on the app anyways
• newsfeed is much longer, more chance of us being published
• much more app centered

The bad

• App looks like garbage… not sure what the hell is going on

I’m sure users are going to sh*t a brick, but they’ll adapt in a month or so. The new site is a blatant ripoff of friend feed. But hey, I like it.

My VPS Link account is up for renewel. Right now I’m paying $25 / mo for:

256 MB Ram
300 GB Bandwith
10 GB storage

I’ve got an ubuntu virtual machine running on Xen. So far everythings worked great.

But so far I’ve been the only user.

As I get closer and closer to launch, that nagging question keeps coming to mind… Will my server handle the load? I know next to nothing about load balancers and server management. I could learn, but I feel this would be a heavy distraction with very little end-user benefit (well, aside from uptime/reliability/ and all that jazz).

The next nagging thought: “well maybe i should switch to a cloud.” I’ve heard great things about Amazon EC2 and S3. Aside from a few recent hiccups, it seems an ideal solution for a small website with (hopefully) high growth potential.

A good resource: 10 Creative Ways to Use Amazon’s Web Service . Most telling:

Webmail, an email hosting company, turned to Amazon Web Services for better reliability as well as cost. They were able to cut their costs by 75% using Amazon’s S3, Simple Queue Service and EC2. Webmail touts Amazon’s services as a shift from

“do-it-yourself to let-the-experts-do-it.”

I’m going to look into this…

Evan Williams, Silicon Valley Entrepreneur, put together a list of the Ten Rules for Web Startups. Top 10 countdowns, ala VH1, have never really been my thing… but coming from the guy who started Twitter I’d imagine it’s pretty legit.

A sampling:

#1: Be Narrow
#2: Be Different
#3: Be Casual
#4: Be Picky
#5: Be User-Centric
#6: Be Self-Centered
#7: Be Greedy
#8: Be Tiny
#9: Be Agile
#10: Be Balanced
#11 (bonus!): Be Wary

I’ve been refining ‘researcher’ pages quite a bit recently. I moved them to the studybob URL rather than the facebook app. I rolled out all kinds of fancy AJAX links, effects, and the like. Pretty soon I’ll implement Scriptaculous for super cool effects.

I’m trying to make the researcher accounts as easy and friendly as possible.