rmurray10 (5) [Avatar] Offline
#1
I just purchased this MEAP and started reading the available chapters. One fundamental component of the book is Opa which does not seem to be a very active project. The github project graphs show very little activity after 2013, the vagrant image (recommended way of installing Opa) has not been updated in two years and breaks when I try to vagrant init mlstate/opa. Do you see the same level of inactivity? If so, are you still committed to Opa and why?

I am very interested in functional programming, concurrent programming, and the new web paradigm you describe. Thanks, Dick Murray
Chris Wilson (1) [Avatar] Offline
#2
I had a very similar question when I was writing the chapter: https://github.com/MLstate/opalang/issues/158

The reply indicated that the authors hadn't abandoned it and were still actively using it. There are a lot of features that I like about it and I think it occupies a nice niche. It also seemed like it fit in with the themes of the book very well, which is why I decided to include it. That being said, with how quickly things move in the JS ecosystem, I'm always on the lookout for something that would be a better fit.
rmurray10 (5) [Avatar] Offline
#3
I fooled around with the vagrant box and got it to run from a local copy of the .box file. It looks like there was a lot of work done before the authors priority shifted. Have you seen http://zderadicka.eu/opa-mixed-impressions/ ? This indicates that the build from sources is fairly straight-forward, which is another path to getting a useful Opa. Your main point about Opa is that it is a single language for both client and server. Have you seen
https://zetafleet.com/blog/2011/03/unified-codebases-with-dojo-node-and-requirejs-the-holy-grail-of-dry-code.html

It has a similar goal of unified code base across client server using a combination of tools. The article was written in 2011, but all three tools (dojo, node and requirejs) seem to be currently active. Requirejs has slowed quite a bit.

It seems that using Opa, however applicable to your thesis, is flawed if the language/runtime is not readily available, making it near impossible to follow the book examples. Your choices seem to be 1) resurrect Opa to the extent that it at least supports chapter 2, or look for an alternative technology even if it is not the perfect fit provided by Opa. I have not gone deeper into (dojo, node, requirejs) as a unified code base, but it may be worth your time to take a brief look.

I realize that the (dojo, node, requrejs) is still lacking a unifying language that is functional and has both client and server in its context. You base some of the transparent web characteristics on a single language, and javascript just does not cut it.

Are you considering any alternatives?

You reference Meteor, which is mainstream but does not get away from javascript. I wonder if Meteor plus casting some of your ideas in ES2015 functional features might be an alternative. Have you seen
https://open.bekk.no/es2015-es6-features-commonly-used-with-functional-style-react

Thanks
516326 (1) [Avatar] Offline
#4
Walmart is rolling out a new price match program called Savings 'walmartone.com' savings catcher login For all the Walmart shoppers out there, this has the potential to save you a bunch of money! This looks like a great new savings option for those of you who use 'www.walmartone.com' for a lot of your shopping. If you shop at other stores 'walmartone login' savings catcher like Kroger and Food Lion, who don't double coupons, you may find this price match program allows you to cut out multiple trips to various stores, while still getting the sale prices.