Forum: JRuby Ruby on mobiles

Announcement (2017-05-07): www.ruby-forum.com is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see rubyonrails.org/community and ruby-lang.org/en/community for other Rails- und Ruby-related community platforms.
Werner Schuster (murphee) (Guest)
on 2008-12-17 04:33
(Received via mailing list)
Seems like XRuby beat JRuby in the race to mobile platforms:
http://www.infoq.com/news/2008/12/rhodes-iphone-wi...


murphee

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
Charles Oliver N. (Guest)
on 2008-12-17 22:59
(Received via mailing list)
Werner Schuster (murphee) wrote:
> Seems like XRuby beat JRuby in the race to mobile platforms:
> http://www.infoq.com/news/2008/12/rhodes-iphone-wi...

And JRuby actually runs Ruby apps. So what's your point?

- Charlie

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
Werner Schuster (murphee) (Guest)
on 2008-12-17 23:14
(Received via mailing list)
Charles Oliver N. wrote:
> Werner Schuster (murphee) wrote:
>> Seems like XRuby beat JRuby in the race to mobile platforms:
>> http://www.infoq.com/news/2008/12/rhodes-iphone-wi...
> And JRuby actually runs Ruby apps. So what's your point?
Code written with the Rhodes platform _are_ Ruby apps.
Also: no, I don't have a point - I was just very surprised to see XRuby
resurface again.

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
Charles Oliver N. (Guest)
on 2008-12-17 23:27
(Received via mailing list)
Werner Schuster (murphee) wrote:
> Charles Oliver N. wrote:
>> Werner Schuster (murphee) wrote:
>>> Seems like XRuby beat JRuby in the race to mobile platforms:
>>> http://www.infoq.com/news/2008/12/rhodes-iphone-wi...
>> And JRuby actually runs Ruby apps. So what's your point?
> Code written with the Rhodes platform _are_ Ruby apps.

Yes, they'll probably work fine, so long as they don't use Ruby features
XRuby doesn't support. Last I checked XRuby's core classes were missing
a lot of functionality, and there were pieces of syntax that just blew
up.

It would certainly be possible to strip down JRuby in the same way; I'm
surprised we were never approached about it.

> Also: no, I don't have a point - I was just very surprised to see XRuby
> resurface again.

You may want to reconsider how you spread news. Posting a message to a
JRuby list saying how XRuby "beat" us to something isn't exactly
friendly. It starts to look like you're trying to drive up hits to your
article, and I'm sure that's not what you intended.

- Charlie

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
Vladimir S. (Guest)
on 2008-12-18 12:38
(Received via mailing list)
Hi Werner,

On Wed, Dec 17, 2008 at 3:32 AM, Werner Schuster (murphee)
<removed_email_address@domain.invalid> wrote:
> Seems like XRuby beat JRuby in the race to mobile platforms:

The re-emergence of XRuby is indeed puzzling. Also, I've been looking
into JRuby in mobile area lately, just to see what our options are.
There was an early prototype of JRuby running on phones a few years
back, presented at JavaOne, it was based on JRuby 1.0.x branch. Since
then, JRuby changed *A LOT*, heavily relying on Java 5+ features.

It seems that the nature of Java environments on the phones is pretty
restrictive (limited resources, limited CPU, no Java 5 features, no
reflection, restricted classloading, etc), so it is a bit challenging
to provide 100% complete Ruby implementation right on the phone, with
no limitations. Looking at the approach you've reported in the
article, those guys also have restrictions like not supporting eval().
Also, looks like they compile ruby code to .class first and then
deliver to the device, right? This is the easier part. Much more
interesting would be to have a full interpreter right on device side,
suitable to run IRB for example. This would be great for quick device
capabilities investigation and for rapid experiments/prototyping.

Having a subset of fulll Ruby implementation is, well, still
interesting, but less practical and less useful from developers point
of view, since developers won't be able to fully use all those
billions of Ruby libraries out there.

So, it seems to me, the time for a full Java-based ruby implementation
is not arrived. Hopefully, in a near(ish) future, we'll see full Java
SE stacks on phones, then running JRuby there would be a trivial
matter. But spending time and effort on overcoming severe restrictions
of aging Java ME CLDC stack would be just not very useful.

Thanks,
  --Vladimir

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
John S. (Guest)
on 2008-12-18 12:50
(Received via mailing list)
I did some initial work on porting the full MRI to Windows Mobile using
cegcc; the existing Ruby CE port was a very, very cut down version
missing a lot of features.

http://www.target-earth.net/gadgets/code/rubyce.html

However the memory requirements proved to be just too tight to get a
fully working stack, though the interpreter does run and I got things
like sockets working, it even got to the point where gem would run and
download remote gem lists.. though unpacking these consume all memory
even on a 192MB+ device.

As a bit of a laugh I even tried running Rails... it actually gets part
way through the startup of the server!

A lot of the problems I found where down to the lack of a relative
directory structure... so the majority of require statements and unit
tests do not work out of the box.

John
---
 Unix & Web Infrastructure Management
 Faculty of Medical Sciences Computing
 University of Newcastle

 Email : removed_email_address@domain.invalid
 Web: http://www.ncl.ac.uk/medev


>> Seems like XRuby beat JRuby in the race to mobile platforms:
>to provide 100% complete Ruby implementation right on the phone, with
>of view, since developers won't be able to fully use all those
>
>---------------------------------------------------------------------
>To unsubscribe from this list, please visit:
>
>    http://xircles.codehaus.org/manage_email
>


---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
Charles Oliver N. (Guest)
on 2008-12-19 11:09
(Received via mailing list)
John S. wrote:
> I did some initial work on porting the full MRI to Windows Mobile using cegcc; the 
existing Ruby CE port was a very, very cut down version missing a lot of features.
>
> http://www.target-earth.net/gadgets/code/rubyce.html
>
> However the memory requirements proved to be just too tight to get a fully working 
stack, though the interpreter does run and I got things like sockets working, it even got 
to the point where gem would run and download remote gem lists.. though unpacking these 
consume all memory even on a 192MB+ device.

That may have improved since they eliminated the bulk index in RubyGems.
Hell, JRuby used to blow its memory cap there for a while...that index
was a terrible load.

> As a bit of a laugh I even tried running Rails... it actually gets part way through the 
startup of the server!
>
> A lot of the problems I found where down to the lack of a relative directory 
structure... so the majority of require statements and unit tests do not work out of the 
box.

We've improved on that somewhat by having require/load and many File
APIs work equally well with URLs like jar:// (and by equally well, I
mean as far as we know they work well). That would probably help solve
the relative pathing logic when running in an embedded device.

I'm not sure what XRuby does here; presumably they have something
similar to load the compiled scripts, since they don't have an
interpreter at all.

- Charlie

---------------------------------------------------------------------
To unsubscribe from this list, please visit:

    http://xircles.codehaus.org/manage_email
This topic is locked and can not be replied to.