Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Creating a Planet Lisp mobile app in 15 minutes with mocl [video] (wukix.com)
56 points by cag_ii on May 14, 2014 | hide | past | favorite | 9 comments


Nice video demo.

I spent some time using mocl last June and wrote about the experience (http://markwatson.com/blog/2013-06/interesting-product-mocl....) - very cool tool.

The sweet spot for mocl would be, I think, apps with relatively simple UIs, but with a lot of functionality that you wanted to implement in Common Lisp. My reason for this opinion is that mocl lets you create both Android and iOS (and now, I think, OS X) apps but the UI is specific to each platform. You would get the most benefit if most of the complexity in your project was non-UI code that could be reused across platforms.


Something I'm toying with the idea of doing is to write a common interface in HTML5 and having that in a webview, then using an evented architecture to communicate with a common lisp backend. This way your logic is native/portable and your UI is portable (although not as fast as native UI most likely).

This saves the pain of having to do some more complicated logic in JS where there's a significant speed penalty (especially in older native browsers) with things like crypto, but still lets you have a common interface/logic.

The only thing you'd have to write for each platform is the glue that sends events between the web view and lisp. I guess it'd be a bit like Phonegap on steroids if I can pull it off.


You can write the UI in Java or Objective C. Two languages, tow APIs.

You could also write much of the UI in Common Lisp. One language, two APIs.

Since you can't reuse much code on the UI side, why not stay in CL as much as possible.


Nice, I used your blog post while I was researching how to get everything up and running in January-ish.

The project is definitely exciting and fun to play with. The on-device repl is showcased well in this video.


I guess I should be happy that there's a (seemingly) profitable CL project. But I wish there was a free "Hobbyist License" (probably with some restriction like not being allowed to publish in app stores).

It looks like a very appealing choice for a side project I'm about to start, and a really fantastic project in general. But I can't justify dropping $200 just to check it out.


Yeah, this looks cool as hell, but I can't justify the expense to mess around with it on my own time. You guys should consider another pricing tier for hobbyist and/or non-commercial use.


How much of CL does it implement? Which of the big libraries do NOT run on mocl?


It implements quite a lot of CL.

But it generates a static application. only a simple eval, no compile, no compile-file, no load, most code information is gone, etc. etc.

If the code makes use of various dynamic features of CL, then it won't run. Thus you have to port/write to a static variant of CL. But that's the purpose: small, static, efficient applications.


There was an issue with https/drakma, but I believe that was fixed in a recent version and you can now use https/drakma. (Previously http only).




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: