I think the major problem is the quick paradigm shift. Take ember. There is a 99% chance that an example on the web is not compatible with the version you are working because the API changes every few minutes (exaggerated *lol*). When you are working with ember you know that the doc you are reading is probably outdated and learn to live with that or move to AngularJS or other frameworks.
In Java there are methods since 1.0, marked as deprecated and nobody ever takes them out of the code because it would break some stuff. That's why Java programming still sucks to all that have tried ruby for a few hours ;) Ruby on the other hand (or Ruby on Rails as framework) handles deprecations rather fast. A "hey, this is deprecated" can change to a "where is it?" in a couple of months. However there are clear migration guides and it stays the same framework. And if bigger things change, they change in modules over years. So you don't have to spend one year to update all your customer programs without earning a penny, which most companies could not afford. What I heard about Angular 2.0 I would not call it Angular 2.0. It is another framework. I think many people will move away from Angular because the switch to Meteor, ember, ... is not producing more work than the switch to Angular 2.0. That's sad, apart from scopes in directives it is really the best framework I know ;) What would be great: Clear commitments from the beginning. If you know how long something will be supported you can make decisions. A clear message about the migration path to the future (throw everything away every x years can be OK if you know it). But the biggest problem is common to all open source stuff: 1 to n people are needing something, they do what they need, other are jumping on that train, not contributing anything (most of them), using everything for free and after some time the people who are driving the development are moving to something different. They can do this any time they want, and it is perfectly OK. What I think is a little bit unfortunate is a "Hey, we have something cool for you. Throw away all you didi in the last years, everything is better now." approach ;) With gems/plugins/components it's simple to choose: Is the component so small and easy that you can maintain it yourself? Use it, otherwise forget it. Writing your own framework is not that easy. But when I am looking in the migration efforts you could be better off using only jQuery in most of your applications. Many people use frameworks like AngularJS because they can do cool things, but most of the things you don't really need. 10% faster development, 5% less bugs for 180% more work (if you have the luck of getting a major Angular update)? Using Angular in one project, definitively will consider other frameworks for future projects... Without a clear message and commitment you can spend 3 - 6 months updating to 2.0 - 1 month after that you can get a "we have something new, 3.0 will be much better, throw it away again". And if they find, that they loose 20, 50, or 80% of the Angular users it could happen that there will be a 1.x fork that outruns Angular 2.0 which can than pass away slowly (not good for the people who wrote all applications new). And even if the user base is splitting and you have 20% 1.x users 20% 2.x users, 20% 3.x users and so on - that does not look perfect for me. In short: the point for me is not what they are doing, but how. Some years ago everybody jumped on open source, now more and more people are moving away, because they calculated that it is more expensive in the long run. Maybe some brilliant minds will show up with some new models of software development in the future ;) -- You received this message because you are subscribed to the Google Groups "AngularJS" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/angular. For more options, visit https://groups.google.com/d/optout.
