Technology Radar May 2013 | Hacker News Reader

Technology Radar May 2013

thoughtworks.com

63 pointsABS16 comments

I'd like to read what specifically lead them to categorize Backbone.js under "hold". They wrote about the various technologies that are recommended but don't provide much of a narrative for any of the "holds".

If you dig back to the last tech radar (http://thoughtworks.fileburst.com/assets/technology-radar-oc...) you'll find the narrative for why Backbone.js was moved to the 'hold' category.

Thanks! Here's the specific section on backbone from page 11:

Backbone.js is a great example of an abstraction pushed too far. While we initially liked the ease of wire-up, in practice it suffers from the same issues as all such databound frameworks from WebForms to client/server tools. We find that it blurs the framework and model too much, forcing either bad architectural decisions or elaborate framework hackery in order to preserve sanity.

They go on to say:

As the industry shifted from desktop GUI development to the web, it seemed natural to port the most successful patterns and designs to the new paradigm. After 15 years of trying, we feel that there are still no component-based frameworks that have successfully achieved this. We recommend not attempting to make web development into something that it fundamentally is not. It is time to accept the page and request-based nature of the web, and focus on the frameworks that support - rather than work against - these concepts.

This seems reasonable, but single page apps don't have a page and request-based nature, and Backbone seems a good fit for this case.

Is it just me or are a large number of their recommendations have no commentary at all?

For example what is "Aggregates as documents" and why is it top?

It is a bit terse. There is a very easy to miss show/hide details link in each section, but it does seem to omit mention of this technique. I think it is talking about using aggregations of data as documents in NoSQL databases. See http://martinfowler.com/bliki/DDD_Aggregate.html and http://martinfowler.com/bliki/AggregateOrientedDatabase.html

Another example, in this context, what are "Big enterprise solutions"?

Always a great read. Strong recommendation for Clojure but just an Assess for Clojurescript. I would have liked to see some specific recommendations for Clojure tooling and frameworks.

Also interesting to see a strong recommendation for Sinatra but no mention at all for Rails. I agree with Sinatra's awesomeness, but was surprised to see noting on Rails.

Because we have so many blips we fade old ones or items that we feel are no longer worth highlighting. We like Rails but most of our readers know that and so we no longer include it on the Radar.

Go is not on their radar yet!

but that would conflict with their own continuous integration system... http://www.thoughtworks-studios.com/go-continuous-delivery

Just a quick reminder: if you only use the popular tools you might miss the perfect weapon for your use case.

Finding the right tools isn't an arranged marriage, it's a continuously evolving affair with twists and turns like a good novel.

Thanks for puttin this together. Always interesting to read what others think is "hot".

Adopt MongoDB? I thought people were starting to realize that it's not so "web scale" after all.

What people are now realizing is that MongoDB is not a one-size-fits-all technology.

When we put something into "Adopt" we feel that its usage patterns and general "fit" are well known. We feel that's the case for Mongo -- it doesn't fit in all cases, but when appropriate it's absolutely a solid go-to platform. More industry awareness around when Mongo is the wrong choice actually makes it easier to adopt.

PostgreSQL for NoSQL

Hell yes, trialling here soon!

Reply on news.ycombinator.com