Other than more people using it and world domination what feature/project would Fantom need most?

I think basic infrastructure is looking pretty good. And with two IDEs, tooling is looking good too. Although I still think you can't have enough tooling.

Other infrastructure level projects might be wrap/re-implement commonly used Java libraries with nice Fantom APIs.

And tutorials, examples, and blog posts are always great.

I remember talking about .Net. How much experience with .Net is needed to finalize the sys so all test pass?

If you have working knowledge of C# and the .NET APIs - you should be able to feel your way around. In most cases the sys APIs are a straight port of the Java implementation.

Biggest impedance is not knowing the .NET APIs - so lots of googling and poking around the (awful) MSDN site (tho think they have improved it since then). So the more you know there, the more it will help you.

Last I checked we were around 98% complete for sys (though might have dropped since we stopped working on .NET full-time)

Low level 3D API

I try to make an OpenGL/WebGL binding. https://bitbucket.org/chunquedong/fan3d But now, it can't work at all.

Why can't it work?

@DanielFath: Sorry, Just some complaints. It's hard to debug the native code.

I think what Fantom needs most is publicity. :)

I would say two things.

One is tutorials/walkthroughs.

It's been the case for a long time that the "examples" section of Docs has been the top resource for figuring out how to write a real Fantom app...and they're not really geared towards that.

It takes a lot of digging, trial and error, and trawling sidewalk to figure out how to set up a production-ready app that actually makes use of Fantom's advantages. It would be great to see a tutorial walking you through getting a basic webapp set up from start to finish, including database access, FWT->JavaScript, the preferred method of service discovery in Fantom (many of us are used to dependency injection), unit testing, etc.

It seems to be a recurring theme on sidewalk that someone will say "why doesn't Fantom support X?" and Andy or Brian will post "it doesn't need to - just do it this other way and it will be better." And you'll be right, but we had no idea that was even the recommended way of doing it because at present there's no (publicly-available) canonical example of what you guys think a production-ready Fantom app should look like, let alone with explanations of why you're doing things the way you are.

Tales has solid tutorials, and it would be great if the core language did too.

The other is testimonials/case studies.

I really wanted to use Fantom for a big project at work but was unable to sell my boss on its being ready for prime time because (A) I couldn't point to very many sites using it and (B) I couldn't quote anyone on why developing in Fantom makes you more productive than other languages.

I was able to sell him on Grails instead, using this page: http://grails.org/testimonials - it would be great if Fantom had something similar we could link to to demonstrate it's ready for the big leagues.

I really, really want my next big project to be written in Fantom. :) Every time something bugs me about a language it now bugs me even more when I think that Fantom does that thing right and yet I'm suck using something else.

IMHO, Fantom needs a bigger development community.... Currently, IMHO, even that fantom is great... its community and most importantly its maintainers are few enough (Brian and Andy) that using it in a large project is very risky from a business perspective... I don't think anyone will use fantom in a large project unless they are ready to contribute to the lang... and they can probably take care of what they need.

What we need is more maintainers and more people using it... more publicity...

In two word: "bigger community"

I think what Fantom needs most is publicity

I agree, more public tutorials, blogs, speaking gigs and talking about Fantom is very important.

I was able to sell him on Grails instead, using this it would be great if Fantom had something similar

This is a good idea, not sure how many we have (several that I know of at least). If you have a project which uses Fantom in production and would like to be included in something like that could you email me - [email protected]?

It takes a lot of digging, trial and error, and trawling sidewalk to figure out how to set up a production-ready app that actually makes use of Fantom's advantages.

This is interesting perspective. We all learn different ways, and so the docs are definitely structured more to how I like to learn - comprehensive details and small focused code samples to see how a given feature works, but leave the overall assembly to reader. But I can definitely see how "putting it all together" is a gap. I'd be interested to hear more about what kind of learning materials others would find useful (and if someone is interested in leading such an effort).

Currently, IMHO, even that fantom is great... its community and most importantly its maintainers are few enough (Brian and Andy) that using it in a large project is very risky from a business perspective

Just a few thoughts on that topic regarding the Fantom core: software supported by big companies versus small companies/teams is a two edged sword. Consider Java 1.6 released in 2006, and we are still waiting for the 1.7 release. That is practically the entire lifetime of Fantom or before the first iPhone was even unveiled. Granted there was the Oracle acquisition of Sun and all, but still 6 six years. Or put another way, who do you think is more likely to grow a platform and respond to your bugs - Andy and I or a big company like Oracle?

That is just for the core, obviously community implies a much bigger realm including all the other supporting projects you might wish to use. In this case the Fantom community definitely needs to expand. But also hope to see a community develop which adheres to small, simple, and elegant software (which would be the anti-Java culture :-)

Don't get me wrong, I am not a java fan... java is too bloated and too big .. and currently is becoming more and more like c++ where it takes more than 10 years for a new version to appear....

I really like fantom, and you are certainly doing a great job... but what I mean is that even the core still needs more people involved from several parties not just your company.. this makes it a lot less risky from a business perspective..

A recent example is mono, when Novel was sold and Attachmate layed off the entire team which maintained the mono core.... The team is trying to recover, but that is not the point, the problem is that a simple decision has destroyed the entire platform... they will continue to provide mono for the mobile and desktop apps (if things goes well)... Now, the problem I am talking about is that my project which based on server mono is practically in the wind waiting to see if mono will make it.... that won't happen with java... it might take years to evolve, but I won't die this way...

Now, Fantom needs to be in between, it needs to have enough maintainers from various parties to make it safe from business perspective and in the same time small enough to evolve fast and not become java.... Maybe something like scala or ruby...

I agree with the tutorials/walkthrough suggestions. The current tutorials are not user-friendly enough for the curious.

FWIW - @peter had started to put together some nice tutorials on the language:

http://learningfantom.blogspot.com/

Some may find the content there useful.

+1 on this: I am finding it extremely difficult to put together an actual development project. Most other languages feature several real world production ready examples that can be used to bootstrap an actual project. I'm finding the process with Fantom to be frustrating enough to consider moving on to something else.

It takes a lot of digging, trial and error, and trawling sidewalk to figure 
out how to set up a production-ready app that actually makes use of Fantom's
advantages. It would be great to see a tutorial walking you through getting 
a basic webapp set up from start to finish, including database access, 
FWT->JavaScript, the preferred method of service discovery in Fantom 
(many of us are used to dependency injection), unit testing, etc.

-doug

Few random thoughts:

1) I totally like the "platform" aspect of Fantom ... but a lot of the people I've introduced to have been put-off by the "runtime" setup ... it really is not much work to get that going, but i guess most (java) devs are used to not have to deal with such thing (or maven do it).

So I do think some sort of click and run installer would be useful (maybe a java webstart package ?)

2) Seems like many people use IntelliJ nowadays, but that Fantom plugin seems abandoned ... I've thought many time of reusing my NB plugin to make an intellij one ... but reuse is limited as each IDE have very different API's .. so it would be a lot of work.

3) Real projects, both as examples and "proofs". There are few out there and skypark is not open source so hard to get a quick idea of what ca be done.

4) Publicity ... well obviously, but really i think the software stack is great, the word needs to get out.

5) Docs are great but i think some sort of "one page" quick start would be useful ... it's all there but kinda spread out across too many pages. It should be very useful to have some sort of "full" example (like a boring address book) ... in a way it's boring, but it's nice to get a quick idea about fantom.

6) Some basic web framework built-in. I understand the lightweight approach of Fantom but it seem setting up a web project (ie to test javascript compilation) is just a bit too much work (setting up the routes and a server) ... everything is there, but i don't know a bit of automation seems missing here. I'm in the opinion that most people nowadays do web projects so that should be very easy .... the new repo / fanr stuff might help here ?

Agree on the "quick start" aspect - I have no problem getting the runtime set up, but I remember in my early days as a developer I was not nearly as comfortable with setting environment variables and the like as I am now.

A quick shell script (or I guess executable for Windows devs) to unpack the files and set environment variables would probably have a very high cost/benefit ratio in terms of what it could do for adoption.

Also on the subject of quick start, in my opinion Kaushik did a great job with Tales (http://www.talesframework.org) - in particular the "Got 10 minutes? Install & Make your first app" is very enticing. Who doesn't have 10 minutes to try something cool and new?

We've definitely tried really hard to have complete and thorough docs, but there seems to be a clear message that we don't have good enough documentation and material to help someone spin up on Fantom from scratch. Either in the form of tutorials or maybe a sample "pet-store" kind of example application. So sounds like that would be an ripe for someone in the community to pitch in!

I think part of the concern is that we don't get what such an app should look like. I know personally whenever I've written a Fantom app, I've felt "this works...but I'm pretty sure there's a better way I could be doing this."

A simple working demo from @brian or @andy that is Doing Things Right, which we could then write a step-by-step tutorial around, would be a huge help in my opinion.

I always felt like sidewalk would be a very good demo/example ... I guess it's being rewritten ?

In particular it uses web mods and probably actors a bit which i think are much more clearly understood with a working example like this.

Sidewalk has indeed been rewritten and is powering http://project-haystack.org and http://skyfoundry.com. We need to add fanr based search and fandoc, and then we will be upgrading this site too. Hoping that happens in Sept/Oct timeframe.

Sidewalk will be a good example and we will open source it hopefully before end of the year. But we made a decision to have sidewalk depend on Folio which is our NoSQL tag database. I am still not sure what we will do there, we might release a simple version of folio as open source or just define an implementation backed by a traditional RDBMS.

In the meantime, it appears that the Tales framework and Spectrum frameworks are good starting points.

And in the Fantom code base we have "real world" apps:

For desktop FWT, Flux is a great example - I have been using that tool almost exclusively since 2008 to write my code

For actors and a real world WebMod, the new Fanr code follows a lot of the design patterns I have been using.

A bit further back in the learning curve, I found the "search" box frustrating when getting started, as I could not get to documentation of terms I wanted to learn about.

For example, if you search for "it block" you get a lot of links to sidewalk topics, and no links into docLang. A search for "it blocks" is better, but it's not clear that the "Closures" link is the one to follow. The best link to show would be: http://fantom.org/doc/docLang/Closures.html#itBlocks

I think an index of terms, or an ordering of the search output in favour of docLang links, would be helpful.

A bit further back in the learning curve, I found the "search" box frustrating when getting started, as I could not get to documentation of terms I wanted to learn about.

The search engine is getting a re-write, and my plan is to show/filter searches by type (docs, code, forum, etc)

Although no "pet-store", but I have written a blog program: slanBlog

I think a ready to use webserver command is one thing that is needed. Which allows you to do something like

fanserver ClientSideAppPod::Main -port 8080 -servlet doFoo MyServer::FooServlet -servlet doFoo2 MyServer::FooServlet2

Added to that a sample pod, which has an index to all the samples showing of the ability to run in the browser.

And very nice would:

Make the flux toolkit work in the browser on this website an allow people to have a workspace on the server after account creation. Then they can test and build there apps on the site and see how nice Fantom is.

When this works nice, make a nice publication and put it on a important news website to pull more attention.

One thing I think might be pretty important to attract attention is some attention to the Fantom brand. That is things like:

• What are the Fantom theme colors?
• What is the official Fantom logo?
• Does Fantom have a mascot? (I remember seeing Andy's sketches on a little phantom mascot).

Even if it has nothing to do with the language itself, good branding will help persuade the pointy haired bosses and Mac users of the world to adopt it.

- What are Fantom theme colors?

The "cobalt" theme on the main page. I've "made" one for Idea.

- What is the official Fantom logo?

I always thought that the browser icon was it's logo.

Yeah that's what I assumed to.

Just borrowed that for my presentation template: http://www.colar.net/fantompres/s5-blank/fantom.html

Then i also made that silly thing, it makes me laugh when i see it, but you might get sued by pacman :)

What are the Fantom theme colors?

Nothing official - but theme for this site is unlikely to change anytime soon - so can't go wrong with any of the current colors - primarily the blues.

What is the official Fantom logo

Likewise nothing official. The heading image is about all we have - but I'm not attached to it. An identifiable logo that could be used without the wordmark would be cool as well. Any bored dribbblers out there?

I remember seeing Andy's sketches on a little phantom mascot.

The "cobalt" theme on the main page. I've "made" one for Idea.

It's not that Fantom is completely unthemed. But it could be stronger.

I think Ruby is the perfect example. The colors are ruby red on a white background. The logo is the ruby gem. They even have it down such that Ruby has its own iconic font.

Fantom does all the technical stuff right. But the aesthetics are just as important to people.

Was that "mascot" made on a beer infused paper napkin? :)

Was that "mascot" made on a beer infused paper napkin? :)

Heh, does sorta look like that - but no, scribbled in my moleskin.