Came across as a weak presentation, though it was interesting nonetheless.

I thought the most interesting thing out of the talk was the Dino RPG. It's great seeing more and more examples of haXe being used out in the field. :)

The problem is that Nicolas has so much stuff to cover in short time. It's simply not possible to be detailed enough to show some of the real jewels and still cover all that ground. But a good overall presentation, nonetheless - showing you the benefits of: type-checked intellisensed browser API's, haXe everywhere and stuff like that.

Une présentation intéressante bien qu'un peu trop rapide ...
Je crois sincèrement qu'il aurait mieux valu couvrir uniquement quelques points essentiels pour éviter l'asphyxie d'une partie de l'assistance .
Au final, l'intérêt de ce genre de conférence n'est pas tant ce que les intéressés ont compris mais bien ce qu'ils ont retenu à la sortie .

En même temps, il est difficile des couvrir un tel sujet en une si courte période, le papa d'HaXe aurait-il eu peur de délaisser une partie de son bébé ? ;)

Allez, bonne continuation .

Une petite critique d'un utilisateur régulier .

I think the talk was very good and I watch a lot of such talks. It was informative and fun and seemed to get very well accepted by the audience.

Very interesting presentation. Although I'm not familiar at all with flash development, it seems to me that haxe currently lacks a framework to structure website development :
It seems to me that Flash usually was "everything on client" and "ask the backend from time to time" and is becoming more on more used for traditionnal Web2.0 application, where almost every single user interaction translates into the "business layer call -> database call -> presentation layer -> refresh the display " cycle (i'm talking MVC framework language there). My problem with is haxe now is : in what way does Haxe helps you organize this mess of stream better ?

If Haxe wants to be used to actually built next gen website (like, gmail or facebook with a flash interface) then It really needs a framework to shine. The war between php frameworks (cakephp, symfony, etc.) themselves, and between framework in general (.Net, ror, python stuff, etc.) weren't so much about the language, but rather on the way they help you organise things between client, server and database. And I'm not talking choosing MVC architecture there but also Data persistence and ORM, form helpers, etc.

I agree that people seem to fall for nice frameworks, but... Basically a language/platform/vm is one thing and on top of it anyone can build frameworks (or other things). I don't think it's Nicholas's job to build frameworks (and I hope he will stay focused on the l/p/v as things are very competitive in that area these days too, people demand advanced programming concepts, performance, concurrency..)

For example, did Guido V. R. make Django, Did authors or Ruby make RoR, same for cakephp, java stuff...

My opinion is as it is also because in my web-dev I haven't found frameworks a way to go for a complex projects. When you have something that is complex and must perform you want to wrap code-concept around the goal and not a goal around some framework. As an example.. how many of out of the box web-apps that get used massively are made with such external frameworks? for example phpBB, phorum, wordpress..

best regards,
Janko

Janko : yes you're right, it's not nicolas's job, but if he wants to be not only a developer but also a product manager, he needs to think about the framework issue as well. Who talked about Ruby before Ruby on Rail (except in Japan) ?

The "one language for frontend and backend" is quite seducing, but i don't think it will be enough to become popular. Using the same language is one thing, but the environment between client side and server side is so different that it doesn't solve the main issue : how do you make the split in your code between what's happening on one and in the other ? And with Flash being able to do a lot more than simply update the color of some div, the possibilities are even bigger, and so is the need for guidelines, best practices and toolbox (aka a framework).

Talking about complex projects, I completely disagree. I don't think a simple complex web project can be performed without a valid framework (either a proprietary or an open source one). A lot of big websites are currently built using ASP.net, RoR or Symfony. Just go to the project's website to see some examples...

The haxe library website is a good start, and I'll take a closer look at it, but I hope everything's going to be tied in one consistent project soon.

Regards,
Benjamin G.

"""Who talked about Ruby before Ruby on Rail (except in Japan) ?"""

Yes I agree about that. RoR was the killer app that made Ruby for what it is now.

Basically I understand your point on the rest of the text too. Some people like frameworks, some like libraries. I generally like library approach but I also use and understand that frameworks are also a way to go.

But, just to philosophise a little. RoR was a hit and pushed Ruby up because it was first RoR like framework. Now there is a lot of ruby like frameworks in many languages so if some lang has it would't IMHO have nearly as effect as Ruby got but it might be a window where a bunch of people who expect this kind of workflow find use in a language.