Decoupling, abstracting and ungluing your applications has been something that people have been talking about for years. Ultimately though, most packages still have dependencies and creating a library-abstract bundle is still hard but it's getting easier. How far are we from having the ability to package re-usable web-focussed widgets that can be composed into applications? What would a world where libraries were all non-framework specific that could be glued in look like? Lets have a look at how far we've come as an industry and what the future could hold.

Comments

Comments are closed.

Brad Bird at 19:47 on 17 Feb 2017

At first I was thinking "are you serious" then half way through I was thinking "he's really not selling it to me." At the end I was like "okay I've actually been doing this all along." I'm a PHP developer that uses Laravel (see what I did there) and we always end up installing 50% more packages than the base install. Very eye opening talk.

Davo Smith at 20:34 on 17 Feb 2017

Good way to finish the conference. Presented will, with great enthusiasm and a clear message.

Gary Fuller at 21:03 on 17 Feb 2017

Agnotism. It's a brave new word. Great way to end conference, with a mix of humour and some (potential) prescience.

Very good talk Michael. It was slightly scary title but then halfway or so I was like "oh wow yeah never thought of it that way".

I understood what you meant with the talk but definitely was not a good idea to inspire a whole community to "NOT use frameworks".
You should focus more on how to write good libraries, instead of focusing so much on your "framework less concept".
Frameworks are the same as workflows, they serve to have standards and structure, please don't inspire people not to use frameworks, inspire them to write good libraries :)
Would suggest have some practical examples of how good libraries should be written, and you view on good practices and benefits of doing so.
Show the path to follow, not where you should not go.

Chris Sherry at 18:52 on 18 Feb 2017

Great talk, with a clever twist. I agree with some of points Bernardo has made here but I also feel that attempting to talk about anything in the future is a hard thing to talk about and I think Michael did a really good job in presenting an interested look back and at a potential future.

Fun speaker, though he made me feel very old if he wasn't even alive at the time the first PHP (pseudo-) frameworks were being released. Also I'm pretty sure "agnotism" isn't a proper word! The general concept was very worthy though - more power to the FIG's elbow, I say.