>
> Which confirms what I've always thought: Dalvik has no essential technical 
>  
>
> reason, but work around licensing issues. You might disagree, but then you 
>  
> must admit that Oracle's CEO has no clue about their own technology. Quite 
>  
> hard to imagine.
>
First of all, as brilliant as Page might be, we should 
not necessarily assume he is a VM engineer. Secondly, I think we must 
assume things a just a tab bit more complicated than whether or not, a Sun 
JVM was capable of solving the problem. Facts:

- Sun had 3 discrete Java runtime stacks, none of which fit the desired 
"Android level" runtime while Sun did not allow partial implementations.
- Google's own court material proves on page 21, how "Cost isn't an issue, 
open sourcing the jVM is".
- Google must have been concerned about the trouble of a committee. In 
fact, page 23 Oracle's header says "Google, Unwilling To Share Control With 
Sun..." which is a strong hint that Sun wanting their say even if it was 
Google's baby.

So it's possible that Google technically could've used a hybrid of Hotspot, 
but that's not the same as saying Dalvik is a workaround of a licensing 
issue or that it doesn't have other merits (I.e. trace-based JIT's and 
registred based VM's are assumed more efficient than method based JIT's and 
stack based VM's).

 "One August 2010 email from Google engineer Tim Lindholm to Rubin mentions 
 

> being asked by Page and Google's other co-founder, Sergey Brin, to review  
> possible alternatives to Java. Lindholm advised Rubin all the other  
> choices ****"suck"**** and urged him to negotiate a license for Java."
>
However, the King of Android (Rubin), advised Page way back in 2005 (page 
22) that C# might be an alternative (which, unlike Java, is open spec 
rather than open source). We can only assume that Lindholm talked Rubin out 
of this idea. Perhaps Google feared the prospect of upsetting Microsoft 
greater than over upsetting Sun, although I think the more likely cause is 
Google wanting to bootstrap off the existing Java community, which they 
have traditionally been aligned with, contributed to and hired people from.

> So, Java is not that poor thing that many would like us to think; at least 
>  
> in Google's thought. According to their engineering, it was clearly the  
> best technical solution around for making Android, and when an alternate  
> solution has been picked merely for licensing issues, they pursued similar 
>  
> solutions and even picked the same language.
>
I'm confused, when you now say "Java", do you then refer to the 3 official 
Sun runtimes (JME/JSE/JEE) or just the language as a medium of expression? 
You imply that Google think Java is bad, yet Google based Android on Java 
so I am not really sure what you are trying to say here. I guess maybe you 
mean to say that a subset of the JSE could've been used instead. Perhaps 
you are right, I don't think we will ever find out, since Sun/Oracle have 
shown little interest in creating a modern mobile platform.
 

> Note that all my comments are related to technology evaluation, and I'm  
> not talking about Google stealing anything.
>
Admittedly, I'm having a hard time seeing actual technological arguments 
here. And I, for one, am happy to not have PermGen JVM freezes in my phone 
all the time. ;) 

-- 
You received this message because you are subscribed to the Google Groups "The 
Java Posse" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/javaposse/-/3b8ONSxMkxQJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/javaposse?hl=en.

Reply via email to