harmony-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paulex Yang <paulex.y...@gmail.com>
Subject Re: [general] longer term roadmap
Date Tue, 28 Nov 2006 18:01:42 GMT
Geir Magnusson Jr. wrote:
>
>
> Paulex Yang wrote:
>> Geir Magnusson Jr. wrote:
>>>
>>>
>>> Paulex Yang wrote:
>>>> Geir Magnusson Jr. wrote:
>>>>>
>>>>>
>>>>> Paulex Yang wrote:
>>>>>> Mikhail Loenko wrote:
>>>>>>> I'm ok with that timeline. We probably need goals for "dotted"
>>>>>>> releases also (like 5.1 where we can improve for example 
>>>>>>> performance)
>>>>>> I'm not sure, but I thought we can only release 5.0_01 if Sun 
>>>>>> didn't have 5.1?
>>>>>
>>>>> Why not?
>>>>>
>>>>> There is a version of the spec, and a version of the 
>>>>> implementation, which are different.
>>>> If so, I prefer to Harmony 1.0(SE 5.0 compatible), 1.1(SE 5.0 
>>>> compatible with performance enhancement) just like what Geronmo 
>>>> does, rather than 5.0, 5.1, which is confusing.
>>>
>>> I think we should consider 5.0, 5.1 simply for marketing reasons and 
>>> clarity.
>>>
>>> If I tell you that I'm running my program under "Harmony 5.1.2", you 
>>> know exactly what spec version of Java I'm using, so you don't have 
>>> to do the math to figure it out.  Yeah, it's easy for 5, but once we 
>>> have 7, that will have things like "Harmony 3.2.1" ?
>>>
>>> I think it might make it easier for users.
>> But what if Sun ships Java SE 5.1 someday? if Harmony already has 
>> 5.1.2 which actually is Java SE 5.0 compatible, what should we do? It 
>> may be not likely of Java 5, but possible for Java 6, 7...
>
> If Sun ships a Java 5.1
>
> a) I'll be really surprised
>
> b) I will get over it and not care.
>
> Sun ships an implementation of the spec.  Yes, it's the RI, but the 
> version of their implementation is meaningful *to them*, because it 
> reflects things like bug fixes in their implementation, performance 
> improvements, etc.
>
> IOW, comparing Harmony vX.Y.Z and Sun vX.Y.Z is meaningless, and 
> there's no way we can ever provide meaning.
>
> However, if you are worried about that. I'm happy with
>
>   <main_spec_ver>.<minor_spec_ver>.<impl_patchlevel>
>
> with minor_spec_ver being 0 always :)
>
>>
>> My point is: this is confusing, either we keep the Java SE spec 
>> version intact like 5.0_12, or we use totally different version 
>> number, like Geronimo, 1.0->J2EE 1.4.  I cannot see why Harmony 5.1.2 
>> is more clear than Harmony 5.0_12?
>
> Why do you consider it a spec version?  There have been 7 updates to 
> Sun's release of Java 5 : 1.5.0_01 through 1.5.0_08 and 2 updates to 
> the java SE 5 spec.  Clearly the updates to the binary are not spec 
> updates, but implementation updates.
I'm not so sure just like most users, that's why I think it is more 
clear to keep "5.0" :).
>
> How does IBM name their JRE releases?
This is output of IBM JDK on my machine, "1.5.0" is kept, and the patch 
updates level are called "Service Refresh 2" (SR2)

 > java -version:
java version "1.5.0"
Java(TM) 2 Runtime Environment, Standard Edition (build 
pwi32dev-20060511 (SR2))
IBM J9 VM (build 2.3, J2RE 1.5.0 IBM J9 2.3 Windows XP x86-32 
j9vmwi3223-20060504 (JIT enabled)
J9VM - 20060501_06428_lHdSMR
JIT  - 20060428_1800_r8
GC   - 20060501_AA)
JCL  - 20060511a

>
> geir
>


-- 
Paulex Yang
China Software Development Lab
IBM



Mime
View raw message