Pusher Raises $1M From Heroku Founders And More To Bring Realtime Tech To Apps | Hacker News Reader

Pusher Raises $1M From Heroku Founders And More To Bring Realtime Tech To Apps

techcrunch.com

29 pointsdchs18 comments

I guess that real-time has lost its original meaning in computing.

http://en.wikipedia.org/wiki/Real-time_computing

Real-time has always meant the use Pusher has as well, even in computing. I've definitely been on "real time" simulations. They also satisfy the "real time" in the other definition you're talking about, as you have to do all the activity in a relatively fast time span with a low average tick duration and mostly constant time constraint.

http://en.wikipedia.org/wiki/Real_time_(media)

>In a real-time computer game or simulation, events in the game occur at the same rate as the events which are being depicted. For instance, in a real-time combat game, in one hour of play the game depicts one hour of combat. Prince of Persia Animal Crossing Nintendogs Dead Space (except for finishing and beginning a new chapter) Night Trap The Last Express

Fantastic. Pusher is an awesome product that eliminates the headaches associated with building a real-time product. I'll never build a chat with polling again. I use it for my website and haven't had any problems (well one, but it was mostly caused by my inexperience with rails)

The best part is the pricing, especially for startups. 20 concurrent users for free is a big number. Compare it to something like MailGun, and it seems like $20/month gets you a lot more value.

Quick question: for someone or a team already extremely well versed in Node.JS, HTML5 and Socket.io, where the core product is already real-time by design from the ground up (rather than a legacy HTML page-based experience that needs to be "up-realtime-d quickly") -- should they be looking into Pusher and how would it help them?

I really mean folks who diligently code their own server sides and client sides "completely" (utilizing OS libs of course) by themselves and work on a for-local-installation product that should not require www/online connectivity (think non-public intranet servers) to run its core real-time functions -- should they be looking at Pusher at all? Or is this "just a quick external JS lib to add a chat, drawing or puzzle game to your popular HTML 4 website"?

I think you can view pusher as being the quick and easy way to add chat and realtime. I would look into Juggernaut which is basically a ruby wrapper for socket.io which runs on node.js and gives you most of the same functionality that you get with pusher, except that you run the server yourself of course.

I've used Pusher on a couple apps and it was extremely simple and painless. However, I've found that it's not that much more difficult to roll my own real-time solution using juggernaut/socket.io/node.js. What I lose is the support, but what I gain is obviously the cost savings.

Pusher seems very expensive for the number of connections supported per tier.

I think the number of connections is more than enough, keep in mind that its concurrent connections. If you're using so many connections where their price seems like a sticking point you should probably roll your own.

Check out http://socket.io for the front-end libraries

It's not too hard to roll your own, I wrote my own over the course of a week in 2007 using Flash XMLSockets and Twisted Python as the backend, check here on hints if you need ideas on implementing your own: http://code.google.com/p/aonic/wiki/Examples

> Check out http://socket.io for the front-end libraries

A note of clarification: socket.io is a front-end AND back-end (node.js) package.

To be fair, "more than enough" depends on the problem being addressed. It's trivial to have tens of thousands of concurrent connections on a $20/month VPS. You get 100 concurrent connections on PusherApp for $20/month. That's a huge difference. Sure, it just works and you do no administration, etc. PusherApp seems like a convenience service for startups with money to burn.

I see the founders are also doing http://new-bamboo.co.uk/

if Pusher got funding, does that mean they have to leave new bamboo and panda?

As far as I'm aware (I'm friends with the founders) they have hired someone to run New Bamboo (Rails Consultancy) and have also merged their other SaaS product Panda (http://pandastream.com - Video Encoding) with another US company.

So the founders full focus is on the Pusher product. Most, if not nearly all investors tend to want you to put in professional management in a service business and spin off the product to a separate company before investing large amounts of seed capital.

ok, but for external visitors, they are running 3 different companies at the same time:

http://www.pandastream.com/about_us Damien Tanner Co-Founder and CEO of Panda Max Willians CoFounder

http://pusher.com/about Max WIlliams CEO Damien Tanner CMO

http://new-bamboo.co.uk/people Max Williams and Damien Tanner. "Founders". No mention of who is running the company on that page

Hi, I'm Max Williams.

Damien and I own/founded all 3 companies at the moment. The CEO of New Bamboo is called Gwyn Morfey, he runs that company full time. Future plans for Panda are not generally being discussed publicly :)

Pusher is where Damien and I spend 99% of our time.

Max

I work in the same office as the Pusher guys: they seem super bright and highly motivated. Very happy for the team!

Thanks! I don't think we've met, come over and say hello at some point!

Congratulations to Max, Damien and Martyn! Great team of highly intelligent people.

Anyone used both Pusher and NowJS?

Reply on news.ycombinator.com