Longer response to twitter user gr (0x6772)

Longer response to twitter user gr (0x6772)

Postby joe » Sat Aug 23, 2014 6:03 pm

- Hey @Oovium, any chance of string functions soon? (Start simple with split() and length(): we can build most of the rest from mechs.)

- .@Oovium For example, the if operator fails on strings, but the IfBub does what I'd expect:

- @Oovium Maybe pop/push on a string inside TailBub works for string mgmt short-term? Burns the source, so arrays would be better, but harder.

- @Oovium Also, if/when you'd like to open the source up to at least a small group, I'd be interested.

Currently, I'm working on a smallish first part of a large project called Aexels. This first part is basically a cellular automata lab and will contain a subsection of Oovium that will allow users to define their own cellular automata. The app will be free. I have a number of reasons for creating this app. One of which, is to serve as a marketing tool for Oovium itself. Currently, the biggest inhibitor to moving Oovium forward is paying rent / trying not to starve to death. If I can figure out how to deal with those issues in the context of working on Oovium it will allow me to concentrate on Oovium full-time.

The Oovium project started 5 years ago and much has changed with Objective-C and the Cocoa libraries since then. My plan was to modernize the code base. However, with the new Swift programming language it may make more sense to simply convert the entire code base to Swift instead.

In addition, I have a number of lower level modifications to Oovium I would like to tackle concurrent to the rewrite. For example, I would like to flesh out the OO aspects of Oovium. Allowing users to create their own object types and define methods on those types. I also want to create a base Object class that all other classes inherit from (this would solve the if function problem with Strings). Oovium also really needs an Array and Map type (which will solve the split issue).

In general I prefer to work on long term solutions (such as Array data type) as opposed to short term ones such as pop/push strings. However, perhaps I can create a custom build for you in the very near term with a handful of tweaks that you need in order to get past any road blocks you are currently having.

As far as open source; it is a difficult issue. Obviously, I have an infinite amount of work and could really use some help. But, I also don't want to eliminate the possibility that Oovium could pay my rent which would allow me to concentrate exclusively on it. Also, there has been a reasonable amount of clamoring for an android version, so perhaps some sort of hybrid approach is possible. At any rate perhaps we can discuss all this further.
User avatar
Site Admin
Posts: 241
Joined: Thu Jul 16, 2009 3:19 am

Re: Longer response to twitter user gr (0x6772)

Postby 0x6772 » Sat Aug 23, 2014 6:53 pm

I'm sorry to hear that finances are a blocker for you. I don't know, nor need to know, the details, but it strikes me that Oovium's already a good demonstration of the ability to realize ideas, which is a thing I know people who pay for these days. Maybe drop me a line or touch base on LinkedIn (by virtue of this forum registration, you've got my email address).

To the topic at hand, I think you're setting the right goals: an array type is way better than some hacky cut-through arbitrarily treating strings as stacks. I want to play more with the toy, and the sort of things I'm used to playing with (my background is Unix systems administration, especially storage, especially backups) are strings, and I've never been anybody's idea of a mathematician, but making the toy a tool demands doing that the Right way. So please don't waste time on doing some special build just for me.

It's been A While since my BA (yes, arts: swarthmore.edu) in Computer Science, but I think "cellular automata" speaks to a rather limited audience. Maybe the Right limited audience, but still. I don't know the "market" such as it is, so you're in a much better position than I am to tell whether that's a good vector to get a larger audience. Also, maybe that's just a thing you want to have done. Either way: seems like a reasonable course, because a niche market is a great target.

I've played around with Swift a bit, and I think I'm willing to cope with it about as much as I'm willing to cope with Ruby, but it is inherently at odds with producing a cross-platform mobile app. I've never looked at writing software for Android (or Windows mobile) devices, so I don't know what that looks like, but I'm certain that rewriting your codebase in Swift will make it fabulous on iOS devices… but lock it out from anything not sold by Apple. Maybe something along the lines of m4 for Swift to [whatever] would be useful? (Seems like a fun thing to write, and coincides tangentially with a job change I'll be making in October 2014.)

The "write open source software but make money" quandary is one with which I'm familiar. I didn't ever do anything notable for the codebase, but NetBSD was my appetizer to what's become my career, and I worked for Wasabi for a hot minute. It seems like simply making Oovium open source would immediately kill income to little effect, so that's dumb. (Although, users could pay you $10-$30 for the app, or pay Apple $100/year to build your code, so it shouldn't be *that* terrifying.) I think you've set yourself a pretty big goal with Oovium (and derivative apps), and it might not be a bad idea to accept a bit of help. I think there are ways (both technical and legal) to share the code with a limited audience without losing ownership over the project.
Posts: 1
Joined: Sat Aug 23, 2014 6:22 pm

Return to Oovium

Who is online

Users browsing this forum: No registered users and 1 guest