When what you’re building turns into a phenomenon accidentally – you really can’t take too much credit for it’s evolution and must consider yourself more an agent of change.  That’s exactly what the Twitter team is (or was) up until now.  In fact, I’d go out on a limb and say that now that the Twitter folks are riding this bronco of a company, they hardly know what to do with it in it’s current fashion.

Jack Dorsey lends much of his fame to his reign at Twitter – but past success means very little in the valley and many entrepreneurs spend the rest of their lives trying to outdo their first startup which came across as “beginner’s luck”.  Square is a lofty idea and only someone with super grandiose position in the tech world who could generate buzz could launch an idea into the buzz stratosphere.  If not someone then a massive company with dollars to spend on development and marketing and with a stronghold position in an existing market.  I’m talking about Verifone and PAYware.

We’ll have to see who wins in the end (or exactly what winning means) but I suspect that it will take much more than Techcrunch and Mashable articles to get Square adopted by retailers far and wide.  I’ve been surprised before though…

What needs to happen is for PayPal to introduce payment interfaces that are friendly to Webkit browsers (at the very least) – considering iPhone’s dominance and the growing number of Android handsets – both released and expected.

eBay made PayPal the payment solution of choice in peer-to-peer commerce and it’s safe to say that the majority of internet and mobile users have a PayPal account.   The problem is (or was) that PayPal couldn’t be used realistically as a payment provider on mobile devices because the user would be redirected to PayPal.com to make their final transaction.

I suspect that the Open API will be flexible enough that deep integration of PayPal in a mobile app session will now be possible.  In fact, I’m tempted to try  building something like this on my own with the reason being that Android developers are a little bit confused about whether it really is a tall order or not to expect users to buy an app with a Google Checkout – I hardly think so having rarely used my Google Checkout account unless it were the only option available to purchase with.

A deep PayPal integration (either opening a web view inside the app that utilizes the new Open API) should only help with paid app conversion on the Android platform and open the ability for the equivalent of in-app purchase feature that exists in the Apple iPhone SDK.

While on Facebook, I noticed a few things that I think are reasons why it is so wildly successful:

[1] Things are so intuitive – from how to add a friend to how to search for them; things just make sense.

[2] The thing feels so polished – yet so “hacked together”.  They are using PHP and they don’t care about hiding it (like a whole bunch of other sites that feel some kind of envy from the Ruby-on-Rails crowd that they have to keep the extension from showing in browser URLs).  They don’t mind having simple querystring’ed URLs to fetch data because it just works.

[3] Things feel so open – yet so closed.  If you’re friends with somebody on Facebook, you get a magnifying glass into their profile.  However, if you don’t know them – you’ve got to earn your way in.  That makes the experience much more intimate than voyeuristic (which is what Myspace is all about).

Okay so that’s just 3 things above but I believe they are worth mentioning – so many sites out there OVER-engineer products and services when, at the end of the day, it all doesn’t matter!  Case in point: Myspace works so well because it’s so stupid and simple.  Who knew people would learn HTML to snazzy up their profile pages?  Myspace themselves didn’t after admitting the HTML “feature” was an accident.

The folks at Twitter recently expressed their difficulties in scaling with RoR – check this and this. They faced moderate backlash from the RoR development community and I consider that to be fairly childish and immature. In general,most members of the community regard the language/platform as the best thing since sliced bread!

The fact is: RoR is relatively new and not enough applications that have grown to the level as Twitter have been built on it to regard it as “enterprise-friendly” – yet. PHP on the other hand – however sloppy it has been in the past – has slowly built itself into a super fast, rich, scalable, integration-friendly language that has a large developer following but not as “connected” to the Web 2.0 phenomenon we’re facing today.

Call me a PHP zealot to balance the playing field out but I think both Ruby-on-Rails and PHP have their boons and brays. However, the long of the short is: RoR is young, sexy, standardized and lucky to be married to Web 2.0 while PHP extendable, flexible, and with a bit more freedom.  With Ruby – Rails is the framework you use.  With PHP, you’ve got a handful to choose from all backed by their own development communities.

We’re seeing the same thing happen with Javascript frameworks (Prototype, Dojo, jQuery, YUI, etc) and with platforms too (OpenLazlo, Apollo, Etelos, and the soon-to-be-launched BungeeLabs).