Bushido (YC S11): An App Store For The Web That Can Kickstart Your Side Projects
techcrunch.com
techcrunch.com
I can take Jason Kincaid's word for it and sign up, or Bushido can actually tell me something, anything, about what it is.
In the meantime, we're launching Bushido as a really simple way to launch Open-source rails apps with one click, a unified log-in system, and premium hosting. We do a revenue share with the developers, so all you do is write the app, put up a link that has the url to your rails' app git repo, which branch to deploy from, and any environmental variables you want the app to launch with (so you can customize it at launch time), and then anyone can instantly launch the app. If they end up converting to premium hosting (based on a few factors like number of allowed user signins and apps), we'll split the revenue with you.
The goal is to help developers be more and more independent working on their projects of passion, and to app users way more choice in the apps they can use.
Let me know if you have any more questions, s@gobushido.com!
I can take Jason Kincaid's word for it and sign up, or Bushido can actually tell me something, anything, about what it is.
Jason's article does indicate they could be on to something interesting, though.
One thing that's been on my mind lately is help desk software. The current leaders are way overpriced IMO ($9 to $49 per month per agent–are you serious?) [1], especially when it's best for everyone at a startup to wear the support hat now and then. Not to mention the fact that some lock in your data.
We once used a home-built ticket system for support (a Rails 2.3.x app that's available at https://github.com/bigfolio/big-help). I'm seriously considering re-writing this app in Rails 3.1 and putting it up on Bushido. The biggest feature request I've seen is the ability to handle incoming emails as tickets, and that's now easy with MailGun or SendGrid.
All that to say I think Bushido could really disrupt some industries (even recent ones) that are over-priced or handcuff customers with arbitrary limits.
---
[1]
http://uservoice.com/plans http://www.zendesk.com/signup http://www.assistly.com/pricing http://www.freshdesk.com/pricing.html
Each app that's deployed has a bunch of resources automatically provisioned for it, so as a developer, you never have to wonder if something's present - the goal is to make the developer omnipotent, so you can do whatever you're thinking of, without thinking about it too much :)
Feel free to shoot me an email at s@gobushido.com, we'd really love to talk to you about the help-desk app (even we're looking for a good one)!
Edit: Forgot to add: the incoming/outgoing email stuff is powered by Mailgun actually - those guys are so awesome. With the wrapper we add though, and because we're very language-focused right now, we can make the api's a joy to use, beyond what even they're able to do currently.
I'm assuming these services are targeting companies with people working full time on support, where that $9 to $49 is a drop in a bucket compared to their salary.
I don't discount the value extracted but to me it reeks of enterprise, per-seat licensing a bit.
I just think an OS, unlimited agent option would be nice for small startups.
Like I said, definitely contact me, I think there could be something very interesting for everyone here :)
I've been meaning to do a blog post on such, but an average single support request costs us around $8. So if you've got a business where the average number of support requests per user are dramatically less than 1.0 per user per month, then it can work out, but I suspect there's a proportional support load to those SaaS systems that you're looking at where it's closer to, say, 0.5 per seat. If you lower the price beyond $9/month, then all of the sudden your profit evaporates.
It's non-obvious until you run the numbers, but it goes something like this:
Support Person Salary $60,000.00
Fully Loaded Employee Cost $90,000.00
Per Hour $46.88
Support Requests / Hour 6
Cost / Request $7.81
That's actually why we removed our $9/month plan. It was just hard to get excited about helping someone out when you realized after the first mail you were going into the hole for such.The interesting thing for where Bushido fits into that is that in some ways it'd be pushing things towards utility pricing (the same way a lot of the cloud sort of stuff has done) -- i.e. the cost of supporting a software stack would be proportional to the cost you generate further up the chain, rather than based on some metric of an "average customer".
They didn't exactly show me the company budget, but I'm as capable of counting engineers as the next guy. Suffice it to say we paid way more than $75k a month for the productivity gains of that custom app. (It was my first encounter with Enterprise Software (TM) and I think I went into programming in part so that no one should ever again have to suffer like us CSRs suffered...)
• No, not unless you have no competitors. And if you have a generic solution to lower support costs by 10%, if you don't have competitors now, you will soon.
• No, software pricing isn't rational. Oh, how I wish it were at times. But in general, unless you're an amazing salesman, your prices have to be at a level that to the customer feel "about right".
You're also talking $20,000 a month in direct costs if you're paying that engineer and designer market wages, or about $10,000 if you're paying them its-not-exploitation-if-you-have-equity-wages. If next to these numbers you think $100 for software is a lot of money, you are not a desirable client.
I think it's less about "a lot of money" and more about there being plenty of room for a competitor. Especially one with a "$49/mo for unlimited agents" model.
And yeah, there are a handful of SaaS products that are open-sourced, but most aren't maintained to a high-enough quality for our tastes. We'll work with developers to up the standard.
Example: Mockr (https://github.com/causes/mockr-old). A sweet app that lets you share and review designs among teams (and as per last check, it was unmaintained). We ported it to Bushido with a few extra lines of code and it's here https://github.com/HashNuke/mockr
If you have apps like these, that you think others would find useful, mail me (akash@gobushido.com) and I'll help you with porting it to Bushido. And if you are using Devise for auth, we have a plugin and it's 5 lines of code promise! Most apps that we've ported, we've ensured that they work fine outside of Bushido too (opensource is giving back. right? :)
I hope you guys succeed, because I would love to integrate with such a platform.
You don't have to wait. Just get back to me at akash@gobushido.com and I'll help you port your Rails apps to Bushido. Bonus points if you are using Devise coz we have a plugin for it that makes it so much easier.
It's lame, I know, but better safe than sorry.
It also sounds like a recipe for bad-blood in the open source world. It emphasizes owning a repo rather than contributing to one that already exists.
It also seems to push for over utilization of resources. Once you get to say 1-10K installs a multi-tenant solution would probably make more sense. What happens then? Does bushido help you roll your customers into one big app? Even if it's not good for bushido's bottom line?
Don't get me wrong, I see a lot of good and interesting things coming from this. I just also see some sticky situations.
Something that's been under discussion for a while at Bushido.
Does bushido help you roll your customers into one big app?
No. Each instance of an app runs in it's own sandboxed environment. That helps you keep your app simple and not worry about scaling for those 10k users. If you want a simpler perspective to look at the service - it sort of provides web apps like mobile apps. Separate instances like how they run on your phone. So tomorrow if there's an online editor app (like Google Docs. Just saying...) on Bushido and you have a Chromebook-like device, you can launch your own instance and that'll all be yours like a separate app. Once there's support for more platforms it'll be easier to see it as "app store for the web". All apps on the web and all you do is launch an instance of apps for yourself.
P.S: I work for Bushido
I'm about to launch something (hopefully going live tonight, announce tomorrow) with a similar philosophy - the economic engine of Open Source needs some more octane!
Hope this takes off, I hope I can partner with these guys in the future.
Seriously, this is something that can have the same effect on application as the web had on publishing. "Just write it and place it out there ...".