Yeah, I for one think this was equally flawed. People have made succesful and quickly iteratable web services/apps in all kinds of languages, including Perl and PHP.
Plus, one single data source like pg had is never that accurate, plus the fact he and Martin were already Lisp guru s helped them in their use of it.
Spend months rewriting the most beautiful backend and nobody care, redesign a button and everyone is excited.
In two cases, the guys running it knew it was going to fail from day one and their business model was to do this in two year chunks, syphon the cash out of the VCs after talking the product up, live the high life and disappear for a bit.
I felt no shame working for them back then but I do now.
Even rubbish code, in any language, can survive for a lot longer by shifting the spotlight of scrutiny to the biggest bottleneck, the database. Which will also be the deciding factor in reducing growing pains.
We had a great time. Free snacks, lot's of parties, luxurious office furniture, skateboarding in the hall... In the end, the company ran out of money before the product reached a useful state.
Good times. It has been some time since then and I have a "normal" job now. Last thing I heard about the founder is that he started a new vc backed company destined to run circles around something.