[waseemsadiq]

I already built this back in 2009

http://techcrunch.com/2009/12/01/inbox2-one-inbox-and-commun...

And here is the desktop client which is now open-source: http://www.techshout.com/img/inbox2.jpg

People do not need this and will not leave their existing email clients because they simply do not see email as being 'stagnated'. It works and gets the job done.

There are lots of problems to fix around email, but a new inbox interface is not one of them.

My advice to you would be to go build on top of for example gmail. There are massive number of things that suck horribly. Yes, search is one of them.

Inbox navigation is another one, but keep in mind that there is a reason why the current line based inbox interfaces work.

The third one is attachments, but that is one I am already tackling with my new startup www.fileboard.com

[GnomeChomsky]

Did you even read the article, or did you just come here to toot your own horn? All of your "advice" is exactly addressed by Fluent, e.g. built on top of Gmail, better search, better inbox navigation, and better attachment handling.

I have no idea how successful Fluent will be at all those things, but it's just peculiar when a site says "we're tackling problems x, y, and z" and you come along and say, "your ideas suck, try addressing problems x, y, and z instead."

[waseemsadiq]

Lol @ thinking that IMAP and gmail are the same thing

Having spent the last 3 years of my life tackling this problem I would say I am pretty qualified to say this solution sucks.

And they are totally not solving the problems I mentioned, they are building a new mail client. Which tends to be a completely different thing.

Anyways just my 0.02$

[waseemsadiq]

Oh and I forgot to mention this is horribly expensive to scale.

We calculated that it cost us about 1.20 euro do keep this up per user per month. And mind you these costs were primarily network bound so it doesn't matter if you have faster hardware.

There are some optimization techniques you can do to bring this down (for example use imap push) but with gmail you are looking at about 2K connections per IP, you do the math.

Actually you can do this (and we are doing this through partners) on the cheap but the experience you can provide will not be appealing unless it is an add-on experience on an existing product.

[finnh]

Can you elaborate on the "2K connections per IP" part?

Do you mean that google only allows you 2,000 concurrent IMAP connections per client IP address, so you need to scale up your requesting IP pool along with your user count?