Interesting criticisms of PB's approach to building gmail.
Can you share a link to a product you've shipped in the past 12 months and some of your experiences about building it?
Is that an appeal to authority I hear? I haven't worked on a product that has shipped in the last 12 months but I have worked with several programming languages and have read some Dijkstra and have done some literate programming, and have gone to university/college for a few years now.
My criticism was based on the stuff PB said. All of that functionality could have been done in a single version but PB didn't want to waste company resources unless he could offer up proof that they were being used wisely.
There's nothing wrong with that, but to say that it's a good development model is awful. How long did GMail stay in beta for? It never really shipped.
GMail has 100+ million users[1]. Imagine how many they could have if they actually did ship something. Crappy development model or not, does it really matter? The thing is widely successful and any insight into how it came about is valuable as a learning experience.
