Is creating our own front-end JS framework still credible in 2015

I’m working as an front-end developer at a large company. We
are thinking about creating our own framework (instead of using angular/react/…) but is it credible to do
this?

The applications we build are mostly 1 page html applications that
get their data from a REST service, local JSON files or a simple web
service that feeds us a JSON file. Think of google map applications,
html forms & single page shops.

Currently we are using jquery and handlebars to handle events and minimalistic templating.

The question

Should we create our own framework? So i have listed my concerns and benefits and wanted to know your view on this:

Concerns

  • can we be as good as the big frameworks?
  • do you need to be a js guru to roll your own framework?
  • are we not always going to be behind the game?
  • will our framework feel old very quickly if you look at more complex frameworks (react, angular)
  • will we constantly have to convince ourselves that we are on the right path

Benefits

  • more in touch with browser technologies
  • technologie investigation is on our side
  • total control over features, syntax & documentation
  • easier to teach to others
  • less need for advanced developers

What benefits or drawbacks am i still missing?

1 Like

I hate this to be the first post, because I really have nothing else to add and am interested in hearing other’s opinions. But…

Why do you see this as a point under Benefits and not under Cons?

I’m thinking more on the backend side of things, but one of the Benefits to using someone else’s framework, especially widely adopted ones, is the wealth of documentation and tutorials. This makes it far easier for someone new to come in and get up to speed very quickly, rather than learning a ton of custom in-house environment stuff and learning all the things they did, instead of doing it some other way they may be more familiar with. Not to mention, the ability to hire someone who is already experienced in a framework and only need to learn the way it was implemented.

Hi, thx for your answer. I have put this under ‘Benefits’ because we are planning to create very good documentation. Using JSDOC and tutorials.

Then if this is the case, what’s stopping you from releasing it to the public as well? This is probably one of the most important things behind the success of pretty much any framework. (and ofc the other obvious factors like how well it works, ease of use, unit testing, etc)

I mean if you have the resources to be able to build and support all of this, then by all means do it. Resources are usually the biggest factor to why people opt to using someone else’s framework. Most people don’t have the time or desire to reinvent the wheel in a slightly different way, nor do they have the time to properly document it for the people coming after them. This can end up being a full time job for a rather large team of developers.

I imagine that teaching users to learn your framework will take a while. Also imagine how much time people will waste of your companies time trying to find stuff in the documentation. This wouldn’t happen if you used a popular framework since it’s easy to find good jquery developers. Also, they will not know how to debug broken code since they are new to it. I dunno, I think you should stick to popular frameworks. You’re just going to make life difficult for yourself.

None of your benefits are so beneficial to warrant this new framework.

Hi, thx… at the moment i may be naive in thinking that we can create a framework that is easier to understand and in return easier to teach to others. Since we have control over the flow of an app and syntax we would have an easier time explaining the framework to less experienced developers.

Same could be said for learning the inside and out of things like Ember or Knockout or React. If you have the resources, why not fork one of the more popular ones and tweak it to do what you want? This is what Open Source is all about.

The time you invest in trying to plan, build, document, teach, maintain, and upgrade your own framework has to be more time than you’d invest in becoming experts on another framework. In my mind, there’s no contest.

It kind of sounds like you’ve got that “itch” :smiley:
I have a mostly functional PHP CMS sitting around in an archive gathering dust somewhere, speaking of that itch…

2 Likes

This topic was automatically closed 91 days after the last reply. New replies are no longer allowed.

Keeping things going with a relevant web comic from today:

http://notinventedhe.re/on/2015-5-19

1 Like