Groovy Now Runs on Android | Hacker News Reader

Groovy Now Runs on Android

infoq.com

52 pointsDanielRibeiro36 comments

It's great the Codehaus implementation of Groovy has finally come to Android, but there's no mention of this on the Groovy users or development mailing lists at http://groovy.329449.n5.nabble.com

Groovy users around the world would appreciate being told first, instead of needing to attend a conference, or follow online rags like infoQ or people's personal blogs. And users subscribed to their developers list should be told in advance what's in the pipeline rather than being sprung with a change of plans this suddenly. We're all expecting Groovy 3 with the meta-object protocol rewrite and Java 8 lambda retrofit to come next, and now we find out on Hacker News of all places that Groovy 2.4 is next.

This has always been a problem with Groovy; terrible communication from the core language developers. It's unfortunate because I think Groovy could be a lot more popular than it is but I think most developers are unaware of its existence even on just the stock JVM.

Well, the fact that I can't recall ANY Groovy core developer answering a comment on HN ever, also speaks volumes. Or even some devoted Groovy contributor.

To be fair, which languages do have their core developers posting here, and especially on a regular basis? Maybe I just don't pay enough attention, but I don't really recall that many. Walter Bright is the primary name that jumps to mind, and I'm actually having trouble coming up with many others.

I've seen Rust contributors comment here, but similarly don't think it's a particularly long list.

Well, I've seen contributors and core team people from .NET (and Azure), Java, Go, Rust, Arc (well, duh), Haskell, Clojure, Julia, Coffeescript and lots of other languages...

Some of them on a regular basis (Rust and Go people especially), others frequently pop up when their languages are discussed.

OK, now that you mention it, I do recall the Rust folks. Some of the others I don't remember, but I'm sure you're right. I probably just don't pay enough attention to that sort of thing. :-)

Well, I as Groovy Core don't follow hacker news normally. There are already enough channels for me to follow and the density of Groovy related themes has been quite low here. It as a thing of trying to spend the little time I have with something useful and being effective about it. So if I answer here, then only because someone pointed me to it and then it is more to try to fight misinformation and FUD

They are mostly Europeans and it is late Saturday night, so I'd cut them some slack. I'll post on Twitter and see if they can respond when the wake up on Sunday.

No, I mean I don't recall seeing them ever.

Two of the despots posted something under their own names 2 yrs ago. I think, though, that one of the Pivotal employees frequently posts under various different usernames.

Guess I am one of the ones from 2 yrs ago, even though I am not aware this has already been 2 yrs. I thought last time was already last year... well, I can be easily mistaken. But I really wonder what those Pivotal accounts are supposed to be. Could be a reason to talk with some people if I had details

You can check chronological comments from a certain user at, say, https://news.ycombinator.com/threads?id=blackdrag or https://news.ycombinator.com/threads?id=glaforge . I was just answering someone else's query when I mentioned that.

Most users on Hacker News use pseudonyms to better avoid the problems possible with having an online presence anywhere, though not all of us can be bothered. I suspect one regular commenter here, though, is associated with the Groovy/Grails developers at Pivotal.

Guillaume Laforge and Jochen Theodorou actually comment a bit. There are probably others whose github and HN user names I can't guess as easily.

It was more popular a few years ago.

Back in 2009, Grails was the big theme in many JUGs in Germany.

We even added Groovy support to our JSF internal SDK as a means to buy into the "has Groovy support" checklist.

Nowadays, if it wasn't for Graddle, I guess it would be just maintenance related projects. At least from what I see as JUG talks.

FWIW, almost everything we are doing at Fogbeam is Groovy (and Grails) based. I will agree that Groovy isn't as "trendy" as it was a couple of years ago, but it still seems to be progressing, and it's worked amazingly well for us. Personally, I'm a huge fan of both Groovy and Grails.

Could you give examples of recent terrible communications from the developers? I use Grails - based on Groovy - and I feel like I've been in the loop.

After posting my comment, Cedric Champeau has since posted an announcement [1] to the users mailing list, giving the reason for the delay: "official Groovy support wasn't decided until GR8Conf, where I wanted to show a prototype". So it seems nowadays only Groovy developers who pay to attend their conferences are polled when making decisions. Those who "only" follow the mailing lists are ignored.

[1] http://groovy.329449.n5.nabble.com/You-can-now-write-Android...

I agree that the mailing list should have been used the same day, but are you seriously unhappy that something is announced on the mailing list with a delay of 4 days, where all of these days had been travel, conference or holiday for any of the core members?

So what's the appropriate communication way for this? I am curious to know how that should have been done

Let's talk about the docs. The Groovy website docs (the "official docs") have some pages that are years out of date.

Let's look at the main tutorial, Groovy in Action. The first edition came out in 2007. In 2009, they began work on the 2nd ed. IT IS STILL NOT AVAILABLE except as an "early release book" [1]

What mystifies me in all this is that Groovy is a language with a company behind it. How can they not have made more than such a feeble attempt to provide current docs?

[1] http://www.manning.com/koenig2/

You might be familiar with last year's Groovy Conundrum in Dr Dobb's magazine...

"For all its strengths, however, Groovy suffered from several important drawbacks: the first and most painful was performance. It was slow — even for a dynamic language. The second limitation was that Groovy had no real design philosophy, save for a wide-ranging pragmatism. Any new feature that brightened enough eyes in the community got thrown in. So the language grew organically in ways that were hard to predict, rather than having a clear philosophy and a long-term trajectory. Finally, there was the problem that I believe most cramped its acceptance: limited documentation. There was one "Bible" and a pair of tutorials that rehashed the basics."

http://www.drdobbs.com/jvm/the-groovy-conundrum/240147731

Have you helped improve the docs? It is an open source project.

When you say improve, you should ask why the docs are broken in the first place.

I wrote a lot of doco for Groovy over 5 yrs ago, but it has a tendency to break between different versions of Groovy. E.g. this doco http://groovy.codehaus.org/JN2515-Closures I wrote stopped working after Groovy 1.5 because the Groovy despots suddenly decided to disallow try statements without a catch or finally clause for Groovy 1.6. I had to go through all my code and add finally{} throughout it all. I put the root problem down to the despots not taking work on the Groovy language spec (JSR-241) seriously.

I'm a Groovy user and I find that the Groovy developers -- who mostly work for Pivotal and host the project on Codehaus -- do a good job of communicating their roadmap and proposed features.

2.4 is a surprise to me. but a pleasant surprise and likely a release enabled by an unexpected breakthrough by Cedric Champeau.

Even if Groovy 3 had its first beta tomorrow, it would take several months for it to become the new 3.0. That has been the case for every Groovy version with a beta and especially for a version with so many changes as 3.0. Also the bigger the changes in 3.0 the longer there will be a Groovy 2.x. You can't expect to have 2.x maintained for, for example a year and have only bugfix versions coming out.

Here is Cedric Champeau's blog post "Groovy on Android" which contains the slides to his presentation: http://melix.github.io/blog/2014/06/grooid.html

Work towards running Groovy on Android has been going on for years, but with this new PR <https://github.com/groovy/groovy-core/pull/436> and the (unofficial?) announcement of Groovy 2.4 it looks like it's finally here.

Champeau didn't say anything about performance. Last year Laforge said it took 20 secs to startup a simple Hello World program [1]. Is performance now better?

[1] http://groovy.codehaus.org/GSoC+2013

I'm an iOS user/developer and Groovy developer that's thinking about dabbling with Android. I haven't tried it yet, but I know they've done a lot of performance work in the last year.

There's an app in the Google Play store that you can try out: https://play.google.com/store/apps/details?id=me.champeau.gr...

I was at Cédric's talk at GR8Conf and he spoke about the performance and the memory footprint. Using @CompileStatic the performance is like a Java native application and the memory consumption is about the same.

I suggest you that keep an eye on the gr8conf web page and see the talk when is published.

I'd be over the moon if Groovy becomes practical for Android development. It would really ease the pain of not having recent versions of Java available.

A lot remains to be seen though - a lot of these Java alternatives turn out to be pretty flaky with weird unsupported corners of functionality, poor integration with IDEs, debuggers, etc., and when you have weird bugs you're in an extremely small isolated minority in trying to figure it out. There's a whole lot more to it than just getting it to compile / run.

I learned a few decades ago that staying out of official supported languages in OS vendors SDKs is opening the doors to extra working hours as you add another layer to the tooling.

So I do it for learning purposes, but for work we just use officially supported languages.

For those interested in other Java alternatives on Android, check out Scaloid: https://github.com/pocorall/scaloid.

I'm really excited to see this. When Android was first released I was over-the-moon thinking about all the various applications I could build. As I started building a few all the noise of Java and the poor testing and tooling was really getting on my nerves.

My most popular release is very event centric which means tons of anonymous inner classes polluting the code and making the entire thing quite aesthetically displeasing.

With Android Studio (IntellJ), Gradle and hopefully Groovy a lot of that noise and pain can be removed. I would be dubious moving over to that or Scala until the Big G approves but it speaks reams for the future.

I does sadden me that Groovy has fallen by the wayside these years (as primarily Groovy developer) given it is such a jump over Java. Scala looks great but for the type of apps (web/mobile) it still is rather verbose and slow to compile - not ideal for RAD imo. With Java 8 coming it's good to see they've taking Lambas over, even if it's a bit clunky.

I wonder how well it runs under ART.

This is really cool.

Reply on news.ycombinator.com