Ask HN: Will you or anyone use this service?

Your target audience are developers. And it's already super easy to do what you want done with pre-built images, yum etc. Then there is a trust issue with someone who's new.

If you want to go ahead and provide a service to setup distributed architectures using Nutch/Hadoop etc, you have an audience, but it's a lot of work and there are people already doing it.

I'm not sure why I'd pay another service for this. What would the added value be?

1. Either you have to host the server on your account, or I have to give you my username and password to my hosting account. This might put people off so just keep that in mind.

2. Also I don't see a reason to limit yourself to cloud platforms. Hosts like linode/dreamhost/etc are pretty popular these days and service like this would help a startup with limited/no knowledge in server platforms.

3. I also wouldn't limit yourself to just Linux. If your familiar with windows systems you might make a little more money on setting up MSSQL, Exchange, and IIS (And if I recall correctly Rackspace now offers Windows in the cloud).

4. Also think about upkeep of the servers as well. If someone is going to pay you to set it up, they will likely pay you to maintain it.

Not knowing anything about the market, there might be opportunity in building large custom clusters instead of just single servers, and use Open Stack as a base. Might make it look like a juicy acquisition for someone.

Just find a way to whitelabel it yourself.

Caveat: you'd have to have ballpark figures on your site. I have this thing about consultants/firms not saying what their rate/price is up front.

The interface needs to be significantly less effort than writing/modifying a Stackscript (http://blog.linode.com/2010/02/09/introducing-stackscripts/)

"Pick this option in your slicehost setup. SCP this file into your home directory. sudo go-get-a-coffee.sh and come back in twenty minutes."

Alternatively, the same thing delivered as a Chef script or what have you. Capistrano and Deprec used to be a fairly good option for getting a generic Rails stack ready on Ubuntu, for example.

As to whether I would pay for this: heck yes, but there is a fairly narrow window for it. When I'm doing green field development, like where I am for a very small window now for Appointment Reminder, I can make extensive changes to my infrastructure without paying ginormous switching costs. After that site goes live for the first time, though, I can't use your service until the next time I launch a product, because I can't trust that you understand my needs intimately enough to handle a migration.

Why would I pay for this? Fairly straightforward: I am not a professional sysadmin -- I am a developer who knows a few things about a few things because he has to. I know I'm going to burn two or three days tweaking my configuration prior to launching AR, even with productivity enhancers like Deprec to use. I will probably make at least a few consequential errors which will lurk until they bring the site down some weeks later, and create technical debt against later upgrades. I do not enjoy configuring servers. Blocking off three days to tweak an Ubuntu install prevents me from writing a two line email to a satisfied client and booking a consulting engagement worth several thousand dollars.

Incidentally: you're going to hear a lot of feedback from developers about how you're gouging them if you charge reasonable prices for this. They are not your customers.

Hypothetically, I would pay for a way to easily grow my single VPS into a multi-machine config.

I'd use your service if it was reasonably priced (and I hadn't invested a lot of time into automating this myself).

The fact that I have taken the time to automate this workflow totally gives your idea merit though.

Make a nice interface where I can select the most common options and allow me to save it for future use. Also make me add a public key, and send me a confirmation email containing an entry for .ssh/config when it's all up and running.

But this is not for everyone neither I think, so I definitely think there is room for your service somewhere in between!

Small point: I would definitely expect such as service to provide passwords in an encrypted manner (as you need to communicate them at the end I guess).

If you are also offering to get a working uncustomized copy of popular software installed (e.g. Wordpress) then that's more interesting. Some bits of software need a few manual tweaks/config changes to get working. This is especially true for when you move away from Apache and to nginx. Nothing rocket sciency, but still time consuming and so could be outsourced.

There are other services (Grid-something, etc.) but last I looked, they were still expensive and had other drawbacks. This needs to be simple, and simpler than most of what I've seen. You need to be able to get onto the boxes if you need, but also manage them with a webapp.

Bonus points: implement security features like virtual networks in my private cloud, so that I can keep my DB servers off of the internet (even though they're technically on the internet). I want to do this easily and without hand-configuring the boxes. Amazon has something like this, but the last time I looked, they were the only ones.

Even with that said, I think that I would normally prefer to vertically integrate that type of work into a startup (incremental adjustments to the deploy script during dev, technical founder uses the deploy script during early days, when the business grows, a dedicated sysadmin will be required anyway) - but that does depend on the price and exactly what is offered.

If you pitched it as: * Our core business is to license common deployment scripts (i.e. the scripts are commercial software, and you sell a license to use them). * For a small fee, you can use our web interface to help you generate a customised version of our script - or you can edit the script yourself. * For a small fee, we can also keep track of all your deployments and deployment scripts, and run your deployment scripts for you on demand via a web interface.

then the value proposition might make it worthwhile if the price was right.

For example, I have a LAMP project I want to run on a VPS and will only access the server over SSH. You could go in and rip out all the crap like FTP servers, mail servers, close unneeded ports and tune it for Apache+PHP and MySQL and I will pay $20-$40 for that.

In 6 months, if my app has taken off I ask if you can update/patch/harden the machine again for $40-60.