AngularJS 1.3: a new release approaches | Hacker News Reader

AngularJS 1.3: a new release approaches

blog.angularjs.org

66 pointswatermel0n39 comments

The worst part of 1.3 (well, anything > 1.2.2) is they're removing comments from their documentation. The docs are already abysmal; now they're removing the only part of the docs that was decent (since people would post solutions to common problems).

Although, to be fair, half the Disqus comments were just complaining about how bad the docs were.

http://blog.angularjs.org/2013/11/farewell-disqus.html

Angular's Disqus comments were a perfect example of the hate and entitlement in open source that makes people quit contributing. While they're not the greatest docs, they're quite passable, especially with the rest of the info on the web (videos explaining concepts, egghead.io, etc). On top of that, it is really easy to contribute. I saw a few small doc issues, fixed them, and they were merged within 6 hours. If people could be more civil and constructive, comments would be fine, but I'm glad they're gone.

> If people could be more civil and constructive, comments would be fine, but I'm glad they're gone.

So it's better to just ignore the problems with the documentation? I barely have time to learn the framework—the lack of good documentation means I delete it from my hard drive. No matter how good a framework is, if it's not documented well, it's a buggy black box.

Even criticism that doesn't offer a solution is better than a lack of feedback. Besides, it's not like Google will get all self conscious and stop contributing because people don't think the documentation is good.

> I barely have time to learn the framework.

If you can't be bothered to deeply study and understand your tools then you probably don't deserve to use them anyway.

duaneb doesn't have much time to learn one particular framework, so you think he doesn't deserve any tools?

Wha?

Yeah totally. And if you can't be bothered to read your car's owner's manual from cover to cover you probably don't deserve to be driving it anyway.

A lack of curiosity is indeed reproachable. And while I do consider knowing how a car works basic common knowledge, I wouldn't advise someone that doesn't have that to start with the manual.

Actually, you should probably read the Chilton (or equivalent) manual cover to cover.

This isn't the context of people using Angular the hammer, it's in the context of people trying to learn Angular, the forklift, and all of the forklift users saying "It's not our fault you can't figure it out!"

> If you can't be bothered to deeply study and understand your tools then you probably don't deserve to use them anyway.

I hadn't even gotten past the evaluation stage. Bad documentation is a huge smell for the rest of the framework—I'm not going to waste my time pursuing a tool that I'm not going to use.

There are plenty of places to offer feedback on Angular, general or specific. All I'm saying is that it's nice to look at the docs and not have to see a big string of "OMG AngularJS docs suck" (only a slight exaggeration).

Also, if your comment is going to be useful on a documentation page, chances are it will be even more useful as a commit to the docs themselves. Mad about doc quality? Whine (give feedback) about it on Twitter, maybe even mention one of the committers.

>Angular's Disqus comments were a perfect example of the hate and entitlement in open source that makes people quit contributing.

This is a perfect example of the entitlement people feel to be free from others' ire or criticism.

If the documentation is so bad. Are there any books available that would help?

Try the following

    - AngularJS http://shop.oreilly.com/product/0636920028055.do
    - Web Application Development With AngularJS http://www.packtpub.com/angularjs-web-application-development/book

Have you read either of them? Don't recommend books you haven't read. I know for a fact that the O'Reilly book is crap.

I am currently reading "Web Application Development With Angular". Its a well written book geared towards who already know angular basics. That's why I suggested O'Reilly book as its introductory which explains angular core concepts.

Another book worth looking at is: Recipes with Angular.js which is covered under Leanpub Unconditional Return Guarantee. so if you don't like the book. 45 days of purchase you can get a 100% refund.

    https://leanpub.com/recipes-with-angular-js

so you can try the above, if you don't like the other two

Sadly I agree that the O'Reilly book is of poor quality. Bad editing, sloppy writing style, terrible organization of the chapters. Whatever happened to O'Reilly's standards?

The Mastering Web Application Development book is pretty good I felt.

Opening up the Angular documentation to have a look, it appears that every page has an "improve this doc" button, which allows you to submit a pull request to get the official documentation changed. Surely if you find that the documentation is unclear, it's better to add a note or a clarification directly into the documentation than write a comment?

>> , which allows you to submit a pull request

well , it would mean you can understand it at first place...

If you work with the K-12 system in the US, you'll discover that a lot of your customers are still happily chugging along with IE8.

And they'll have no reason to change if websites continue to cater to them.

I don't think the K-12 system is going to upgrade because AngularJS is dropping support for IE8. I mean, I don't think it'll be a major factor in their decision. Probably funding will be a big factor. Then AngularJS.

But they will change soon enough. It is just that IE8 is the ceiling of Windows XP, which is end-of-life in 4 months. If that does not convince, the bad press of all the vulnerabilities that follows, will.

Another driver: ipads and chromebooks are taking over.

Speed? Security? Browser improvements?

Also more than a few major US corporations. Some of which can't upgrade easily because of third-party tools that don't support anything newer. Or worse, internally developed software which is still in used, but is hard to upgrade because the original developer moved on, and no one is tasked with maintaining it.

This is my client base at $dayjob. As of last year, the last of them had finally let go of IE6. It would be nice to think our influence had something to do with it, but Microsoft had been trying to get them to ditch it for years, and they have a hell of a lot more pull than anyone else.

Even most big enterprises now are supporting IE 9....

Well, time moves on. If you are developing for clients using IE 8, you probably can live without the latest Angular version too.

I think it is necessary that projects keep focused and that means deciding what not to do.

A few K-12 schools using IE8 is not a compelling reason for a modern Javascript framework to continue supporting it.

For the lazy, Angular 1.3 drops IE8 support. This is a potentially big problem for my company, since we have a big contract that stipulates IE8 support. Maybe this is the final straw that gets the client to drop IE8 support (which would be nice), but I suspect this will stymie us from upgrading further for these particular projects.

nah, but 1.3 does.

Whoops, thanks for that correction!

More particularly, they are stopping automated testing of IE8 in their CI system. However, they specifically note that they are happy for the community to continue IE8 support, and expect those with significant dollars that still need IE8 support to join together and keep up the IE8 support. Maybe your client that needs IE8 support is willing to pay extra to keep it, and you can be a part of the solution!

Contracts can be negotiated and renegotiated.

Dropping IE 8 support is tricky. For those who produce web apps for use in the enterprise arena, IE 8 can be a significant part of your user base. In the last month, 15% of our users accessed our site via IE 8. As such, it may be another year, or so, until we can take a serious look at Angular. That, of course, assumes that twelve months hence they don't drop support for IE 9.

I guess this was inevitable with Microsoft dropping support for Windows XP.

Inevitable and good. I'd much rather the Angular team spend time on features and bugfixes for mainstream browsers than supporting a 54-month-old browser version.

Too soon. There are third party components still struggling to update to 1.2.

I just had to update a bunch of directives for 1.2, so I see where you're coming from, but Angular is still young enough that I'll welcome some more major changes if it makes it a better framework in the long-term. It's frustrating to be an early adopter, but I'm loving it enough to bite the bullet.

I believe I saw some slides a while back about their long-term roadmap which made these changes seem relatively minor. If they get around to everything they want to, Angular will be very different across the board in a few years.

Reply on news.ycombinator.com