30 comments

[ 2.5 ms ] story [ 76.2 ms ] thread
Nice idea as a faster incremental development tool, but not something that one should run in production.
Why not? Stick an HTTP cache (Varnish, Cloudflare, whatever) in front of it, and I see no reason not to do the preprocessing on the fly.

Also: "Harp apps run directly with the built-in web server, or exported to HTML, CSS & JS, putting you in full control of your production environment."

I agree, I'd run it production. Full disclosure: I know the team. But I'm a fan of the approach!
Just because you can, doesn't mean you should.
Again, why not?

It would reduce the number of differences between dev and production, for one.

>"A witty saying proves nothing."

Node + Varnish + (optionally) CDN is a very production-ready stack for serving static content.

It would be cool to support CommonJS modules in the browser using Browserify or similar as well.
Yea, CommonJS would be rad.
Not a solution to a problem I have* but I like the choice to carve out a new niche and experiment with a different mix of features made prominent in the web/app server.

* Unless my assumptions about the control over the caching of various template fragments and rendering work is wrong, which is totally possible.

What are those assumptions? What problem do you have related to caching template fragments and rendering?
I would rather have a boiler plate express project that provides the same thing as this without any of the framework lock in.

Cool thing none the less if you already develop with that exact tech stack.

No Sass support? :(
We started with LESS and Stylus, but Sass is in the works.
This is a pretty similar idea to my project Slinky [0], and it's definitely a productive way to do single page development. Slinky adds source dependency management (a.js depends on b.coffee), a built in proxy, concatenation/minification for final deployment and some other treats.

[0] http://mwylde.github.io/slinky/

It would be cool if they added a package.json file with the necessary dependencies and a start script defined.
I'm using hem, with a heroku hosting, which is doing the same kind of job basically, does harp provide any kind of value added ?
Nice pivot guys, congrats on finally launching!

Weren't you originally supposed to be similar to site44? i.e. Static site hosting using dropbox.

Why the change?

We haven't changed. We are just launching our open source server / APF first, building a little more community around that and then we will launch Harp.io.
I'm curious: did you release the server first because the harp.io bits aren't quite ready yet? Or is building a community for the backend server intended to eventually drive customers to the hosted service?
Honestly, both. We are big believers in open source and creating more value then we can capture. This was true when we created PhoneGap, and still true today.

The Platform is not far off and we have people using it, we just want to apply a little more polish and make the experience exceptional.

So unless I'm missing something, this is more/less Jekyll with generalized capability and less blog-focus? Looks interesting.
organized*, not orginized
Relevant: also check out Backlift (https://www.backlift.com/) -- it includes stuff like pre-processing and includes Dropbox integration + a backend + more.. A potential game changer in terms of dev time.
I know these guys, I really like what they're doing
Today we launched our open source tool, but we also have a hosted platform that is build on Dropbox. See harp.io for that.

We have hung out with the Backlift guys - they are awesome.

Maybe I'm missing something from a design/marketing perspective, but... am I mistaken in thinking that the ice cream imagery totally unrelated to everything else?

On the whole, though, I love the idea. Hope I get a chance to try it out.

So how does this compare to yeoman?
Awesome work! I love the deploy to github pages feature!