1.0.0 release?

[Jim Donofrio <do...@gmail.com>]

Brock,

What is your thought on doing a 1.0.0 release? I still would like to 
finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but maybe 
all new JIRA's could get scheduled for a different version, 1.0.1? or 
1.1.0? I think we have enough new features/bug fixes: mapreduce 
Combiner, custom counter checking, improved passing of the conf, better 
support for java/other serializations, better error messages, deprecated 
string methods, AssertionError's instead of RuntimeException's, separate 
source, binary distributions.

Should the version really jump up to 1.0.0, maybe 0.9.0 would be better. 
There are no real massive changes, we could continue in the 0.10, 0.11, 
etc versions until the new api is done which would become 1.0.0?

Re: 1.0.0 release?

[Brock Noland <br...@cloudera.com>]

I have no idea a maven parrot!  I just copy what I see until it works.

I will update the doc and give that a try for the 0.9 release.

Brock

On Thu, Mar 15, 2012 at 2:41 PM, Jim Donofrio <do...@gmail.com> wrote:
> Also in the Deploy to Staging Repo part, I am unclear why we run both:
>
> mvn clean site verify
> mvn clean deploy -Pdeploy -Pjavadoc
>
>
> Deploy will also run verify so you could probably just run
>
> mvn clean site deploy -Pdeploy -Pjavadoc
>
> otherwise the site wont make it into the docs folder again
>
>
>
> On 03/15/2012 12:08 PM, Brock Noland wrote:
>>
>> Basically we decide to do release and then follow the procedures
>> below. Anyone interested in doing the RM role would benefit from
>> executing the first steps under "First time release managers"
>>
>> https://cwiki.apache.org/confluence/display/MRUNIT/How+to+Release
>>
>> The procedures seem to work well, I have tried to fix any hiccups I
>> have had. I copied them from WHIRR.  However, we do accidentally
>> include a .jar file in our tar.gz which people note every time we do a
>> release. As such, I'd like to get MRUNIT-61 solved beforehand.
>>
>> Brock
>>
>> On Thu, Mar 15, 2012 at 10:56 AM, Jim Donofrio<do...@gmail.com>
>>  wrote:
>>>
>>> Yes you can be the release manager. How does this release process work
>>> since
>>> we may not do the release for a few weeks, it depends how long it takes
>>> to
>>> get through the remaining issues
>>>
>>>
>>> On 03/15/2012 09:18 AM, Brock Noland wrote:
>>>>
>>>> Hi,
>>>>
>>>> I am for a release, good to release early and often! Jarek, I will
>>>> look at your patch again today. It sounds like you guys are in favor
>>>> of 0.9 release. I am as well.
>>>>
>>>> Jim, you are working on MRUNIT-69, so I figure I should be the Release
>>>> Manager? If you, or anyone else, wants to try the RM role, feel free
>>>> to speak up!!
>>>>
>>>> If no one speaks up, I will send out a vote on the release + RM role.
>>>>
>>>> Cheers!
>>>> Brock
>>>>
>>>> On Thu, Mar 15, 2012 at 4:05 AM, Jim Donofrio<do...@gmail.com>
>>>>  wrote:
>>>>>
>>>>> Yes custom counter checking is not in yet but it will be for the
>>>>> 1.0.0/0.9.0
>>>>> release
>>>>>
>>>>>
>>>>> On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
>>>>>>
>>>>>> I'm not Brock nor committer, but I would like submit my feedback
>>>>>> anyway
>>>>>> :-)
>>>>>>
>>>>>> Custom counter checking is still not in. I've submitted last version
>>>>>> of
>>>>>> my
>>>>>> patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting
>>>>>> for
>>>>>> review. I'll be happy to finish the page into committable form.
>>>>>>
>>>>>> I would personally vote for version 0.9.0, but I don't mind using
>>>>>> 1.0.0
>>>>>> either.
>>>>>>
>>>>>> Jarcec
>>>>>>
>>>>>> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>>>>>>>
>>>>>>> Brock,
>>>>>>>
>>>>>>> What is your thought on doing a 1.0.0 release? I still would like to
>>>>>>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>>>>>>> maybe all new JIRA's could get scheduled for a different version,
>>>>>>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>>>>>>> mapreduce Combiner, custom counter checking, improved passing of the
>>>>>>> conf, better support for java/other serializations, better error
>>>>>>> messages, deprecated string methods, AssertionError's instead of
>>>>>>> RuntimeException's, separate source, binary distributions.
>>>>>>>
>>>>>>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>>>>>>> better. There are no real massive changes, we could continue in the
>>>>>>> 0.10, 0.11, etc versions until the new api is done which would
>>>>>>> become 1.0.0?
>>>>
>>>>
>>>>
>>
>>
>



-- 
Apache MRUnit - Unit testing MapReduce - http://incubator.apache.org/mrunit/

Re: 1.0.0 release?

[Jim Donofrio <do...@gmail.com>]

Also in the Deploy to Staging Repo part, I am unclear why we run both:

mvn clean site verify
mvn clean deploy -Pdeploy -Pjavadoc


Deploy will also run verify so you could probably just run

mvn clean site deploy -Pdeploy -Pjavadoc

otherwise the site wont make it into the docs folder again


On 03/15/2012 12:08 PM, Brock Noland wrote:
> Basically we decide to do release and then follow the procedures
> below. Anyone interested in doing the RM role would benefit from
> executing the first steps under "First time release managers"
>
> https://cwiki.apache.org/confluence/display/MRUNIT/How+to+Release
>
> The procedures seem to work well, I have tried to fix any hiccups I
> have had. I copied them from WHIRR.  However, we do accidentally
> include a .jar file in our tar.gz which people note every time we do a
> release. As such, I'd like to get MRUNIT-61 solved beforehand.
>
> Brock
>
> On Thu, Mar 15, 2012 at 10:56 AM, Jim Donofrio<do...@gmail.com>  wrote:
>> Yes you can be the release manager. How does this release process work since
>> we may not do the release for a few weeks, it depends how long it takes to
>> get through the remaining issues
>>
>>
>> On 03/15/2012 09:18 AM, Brock Noland wrote:
>>> Hi,
>>>
>>> I am for a release, good to release early and often! Jarek, I will
>>> look at your patch again today. It sounds like you guys are in favor
>>> of 0.9 release. I am as well.
>>>
>>> Jim, you are working on MRUNIT-69, so I figure I should be the Release
>>> Manager? If you, or anyone else, wants to try the RM role, feel free
>>> to speak up!!
>>>
>>> If no one speaks up, I will send out a vote on the release + RM role.
>>>
>>> Cheers!
>>> Brock
>>>
>>> On Thu, Mar 15, 2012 at 4:05 AM, Jim Donofrio<do...@gmail.com>
>>>   wrote:
>>>> Yes custom counter checking is not in yet but it will be for the
>>>> 1.0.0/0.9.0
>>>> release
>>>>
>>>>
>>>> On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
>>>>> I'm not Brock nor committer, but I would like submit my feedback anyway
>>>>> :-)
>>>>>
>>>>> Custom counter checking is still not in. I've submitted last version of
>>>>> my
>>>>> patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting
>>>>> for
>>>>> review. I'll be happy to finish the page into committable form.
>>>>>
>>>>> I would personally vote for version 0.9.0, but I don't mind using 1.0.0
>>>>> either.
>>>>>
>>>>> Jarcec
>>>>>
>>>>> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>>>>>> Brock,
>>>>>>
>>>>>> What is your thought on doing a 1.0.0 release? I still would like to
>>>>>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>>>>>> maybe all new JIRA's could get scheduled for a different version,
>>>>>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>>>>>> mapreduce Combiner, custom counter checking, improved passing of the
>>>>>> conf, better support for java/other serializations, better error
>>>>>> messages, deprecated string methods, AssertionError's instead of
>>>>>> RuntimeException's, separate source, binary distributions.
>>>>>>
>>>>>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>>>>>> better. There are no real massive changes, we could continue in the
>>>>>> 0.10, 0.11, etc versions until the new api is done which would
>>>>>> become 1.0.0?
>>>
>>>
>
>

Re: 1.0.0 release?

[Jim Donofrio <do...@gmail.com>]

I dont know if you saw MRUNIT-71 but I dont think mvn clean 
assembly:assembly verify is right. assembly is bound to the package 
phase so assembly:assembly is unnecessary. Also you need to call site 
before verify in order to copy the site the docs folder. If we dont want 
to copy the site in, we should get rid of the docs folder in the tar.gz. 
I think the right command would be mvn clean site verify



On 03/15/2012 12:08 PM, Brock Noland wrote:
> Basically we decide to do release and then follow the procedures
> below. Anyone interested in doing the RM role would benefit from
> executing the first steps under "First time release managers"
>
> https://cwiki.apache.org/confluence/display/MRUNIT/How+to+Release
>
> The procedures seem to work well, I have tried to fix any hiccups I
> have had. I copied them from WHIRR.  However, we do accidentally
> include a .jar file in our tar.gz which people note every time we do a
> release. As such, I'd like to get MRUNIT-61 solved beforehand.
>
> Brock
>
> On Thu, Mar 15, 2012 at 10:56 AM, Jim Donofrio<do...@gmail.com>  wrote:
>> Yes you can be the release manager. How does this release process work since
>> we may not do the release for a few weeks, it depends how long it takes to
>> get through the remaining issues
>>
>>
>> On 03/15/2012 09:18 AM, Brock Noland wrote:
>>> Hi,
>>>
>>> I am for a release, good to release early and often! Jarek, I will
>>> look at your patch again today. It sounds like you guys are in favor
>>> of 0.9 release. I am as well.
>>>
>>> Jim, you are working on MRUNIT-69, so I figure I should be the Release
>>> Manager? If you, or anyone else, wants to try the RM role, feel free
>>> to speak up!!
>>>
>>> If no one speaks up, I will send out a vote on the release + RM role.
>>>
>>> Cheers!
>>> Brock
>>>
>>> On Thu, Mar 15, 2012 at 4:05 AM, Jim Donofrio<do...@gmail.com>
>>>   wrote:
>>>> Yes custom counter checking is not in yet but it will be for the
>>>> 1.0.0/0.9.0
>>>> release
>>>>
>>>>
>>>> On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
>>>>> I'm not Brock nor committer, but I would like submit my feedback anyway
>>>>> :-)
>>>>>
>>>>> Custom counter checking is still not in. I've submitted last version of
>>>>> my
>>>>> patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting
>>>>> for
>>>>> review. I'll be happy to finish the page into committable form.
>>>>>
>>>>> I would personally vote for version 0.9.0, but I don't mind using 1.0.0
>>>>> either.
>>>>>
>>>>> Jarcec
>>>>>
>>>>> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>>>>>> Brock,
>>>>>>
>>>>>> What is your thought on doing a 1.0.0 release? I still would like to
>>>>>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>>>>>> maybe all new JIRA's could get scheduled for a different version,
>>>>>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>>>>>> mapreduce Combiner, custom counter checking, improved passing of the
>>>>>> conf, better support for java/other serializations, better error
>>>>>> messages, deprecated string methods, AssertionError's instead of
>>>>>> RuntimeException's, separate source, binary distributions.
>>>>>>
>>>>>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>>>>>> better. There are no real massive changes, we could continue in the
>>>>>> 0.10, 0.11, etc versions until the new api is done which would
>>>>>> become 1.0.0?
>>>
>>>
>
>

Re: 1.0.0 release?

[Brock Noland <br...@cloudera.com>]

Basically we decide to do release and then follow the procedures
below. Anyone interested in doing the RM role would benefit from
executing the first steps under "First time release managers"

https://cwiki.apache.org/confluence/display/MRUNIT/How+to+Release

The procedures seem to work well, I have tried to fix any hiccups I
have had. I copied them from WHIRR.  However, we do accidentally
include a .jar file in our tar.gz which people note every time we do a
release. As such, I'd like to get MRUNIT-61 solved beforehand.

Brock

On Thu, Mar 15, 2012 at 10:56 AM, Jim Donofrio <do...@gmail.com> wrote:
> Yes you can be the release manager. How does this release process work since
> we may not do the release for a few weeks, it depends how long it takes to
> get through the remaining issues
>
>
> On 03/15/2012 09:18 AM, Brock Noland wrote:
>>
>> Hi,
>>
>> I am for a release, good to release early and often! Jarek, I will
>> look at your patch again today. It sounds like you guys are in favor
>> of 0.9 release. I am as well.
>>
>> Jim, you are working on MRUNIT-69, so I figure I should be the Release
>> Manager? If you, or anyone else, wants to try the RM role, feel free
>> to speak up!!
>>
>> If no one speaks up, I will send out a vote on the release + RM role.
>>
>> Cheers!
>> Brock
>>
>> On Thu, Mar 15, 2012 at 4:05 AM, Jim Donofrio<do...@gmail.com>
>>  wrote:
>>>
>>> Yes custom counter checking is not in yet but it will be for the
>>> 1.0.0/0.9.0
>>> release
>>>
>>>
>>> On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
>>>>
>>>> I'm not Brock nor committer, but I would like submit my feedback anyway
>>>> :-)
>>>>
>>>> Custom counter checking is still not in. I've submitted last version of
>>>> my
>>>> patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting
>>>> for
>>>> review. I'll be happy to finish the page into committable form.
>>>>
>>>> I would personally vote for version 0.9.0, but I don't mind using 1.0.0
>>>> either.
>>>>
>>>> Jarcec
>>>>
>>>> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>>>>>
>>>>> Brock,
>>>>>
>>>>> What is your thought on doing a 1.0.0 release? I still would like to
>>>>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>>>>> maybe all new JIRA's could get scheduled for a different version,
>>>>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>>>>> mapreduce Combiner, custom counter checking, improved passing of the
>>>>> conf, better support for java/other serializations, better error
>>>>> messages, deprecated string methods, AssertionError's instead of
>>>>> RuntimeException's, separate source, binary distributions.
>>>>>
>>>>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>>>>> better. There are no real massive changes, we could continue in the
>>>>> 0.10, 0.11, etc versions until the new api is done which would
>>>>> become 1.0.0?
>>
>>
>>
>



-- 
Apache MRUnit - Unit testing MapReduce - http://incubator.apache.org/mrunit/

Re: 1.0.0 release?

[Jim Donofrio <do...@gmail.com>]

Yes you can be the release manager. How does this release process work 
since we may not do the release for a few weeks, it depends how long it 
takes to get through the remaining issues

On 03/15/2012 09:18 AM, Brock Noland wrote:
> Hi,
>
> I am for a release, good to release early and often! Jarek, I will
> look at your patch again today. It sounds like you guys are in favor
> of 0.9 release. I am as well.
>
> Jim, you are working on MRUNIT-69, so I figure I should be the Release
> Manager? If you, or anyone else, wants to try the RM role, feel free
> to speak up!!
>
> If no one speaks up, I will send out a vote on the release + RM role.
>
> Cheers!
> Brock
>
> On Thu, Mar 15, 2012 at 4:05 AM, Jim Donofrio<do...@gmail.com>  wrote:
>> Yes custom counter checking is not in yet but it will be for the 1.0.0/0.9.0
>> release
>>
>>
>> On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
>>> I'm not Brock nor committer, but I would like submit my feedback anyway
>>> :-)
>>>
>>> Custom counter checking is still not in. I've submitted last version of my
>>> patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting for
>>> review. I'll be happy to finish the page into committable form.
>>>
>>> I would personally vote for version 0.9.0, but I don't mind using 1.0.0
>>> either.
>>>
>>> Jarcec
>>>
>>> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>>>> Brock,
>>>>
>>>> What is your thought on doing a 1.0.0 release? I still would like to
>>>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>>>> maybe all new JIRA's could get scheduled for a different version,
>>>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>>>> mapreduce Combiner, custom counter checking, improved passing of the
>>>> conf, better support for java/other serializations, better error
>>>> messages, deprecated string methods, AssertionError's instead of
>>>> RuntimeException's, separate source, binary distributions.
>>>>
>>>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>>>> better. There are no real massive changes, we could continue in the
>>>> 0.10, 0.11, etc versions until the new api is done which would
>>>> become 1.0.0?
>
>

Re: 1.0.0 release?

[Brock Noland <br...@cloudera.com>]

Hi,

I am for a release, good to release early and often! Jarek, I will
look at your patch again today. It sounds like you guys are in favor
of 0.9 release. I am as well.

Jim, you are working on MRUNIT-69, so I figure I should be the Release
Manager? If you, or anyone else, wants to try the RM role, feel free
to speak up!!

If no one speaks up, I will send out a vote on the release + RM role.

Cheers!
Brock

On Thu, Mar 15, 2012 at 4:05 AM, Jim Donofrio <do...@gmail.com> wrote:
> Yes custom counter checking is not in yet but it will be for the 1.0.0/0.9.0
> release
>
>
> On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
>>
>> I'm not Brock nor committer, but I would like submit my feedback anyway
>> :-)
>>
>> Custom counter checking is still not in. I've submitted last version of my
>> patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting for
>> review. I'll be happy to finish the page into committable form.
>>
>> I would personally vote for version 0.9.0, but I don't mind using 1.0.0
>> either.
>>
>> Jarcec
>>
>> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>>>
>>> Brock,
>>>
>>> What is your thought on doing a 1.0.0 release? I still would like to
>>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>>> maybe all new JIRA's could get scheduled for a different version,
>>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>>> mapreduce Combiner, custom counter checking, improved passing of the
>>> conf, better support for java/other serializations, better error
>>> messages, deprecated string methods, AssertionError's instead of
>>> RuntimeException's, separate source, binary distributions.
>>>
>>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>>> better. There are no real massive changes, we could continue in the
>>> 0.10, 0.11, etc versions until the new api is done which would
>>> become 1.0.0?



-- 
Apache MRUnit - Unit testing MapReduce - http://incubator.apache.org/mrunit/

Re: 1.0.0 release?

[Jim Donofrio <do...@gmail.com>]

Yes custom counter checking is not in yet but it will be for the 
1.0.0/0.9.0 release

On 03/15/2012 04:54 AM, Jarek Jarcec Cecho wrote:
> I'm not Brock nor committer, but I would like submit my feedback anyway :-)
>
> Custom counter checking is still not in. I've submitted last version of my patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting for review. I'll be happy to finish the page into committable form.
>
> I would personally vote for version 0.9.0, but I don't mind using 1.0.0 either.
>
> Jarcec
>
> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>> Brock,
>>
>> What is your thought on doing a 1.0.0 release? I still would like to
>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>> maybe all new JIRA's could get scheduled for a different version,
>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>> mapreduce Combiner, custom counter checking, improved passing of the
>> conf, better support for java/other serializations, better error
>> messages, deprecated string methods, AssertionError's instead of
>> RuntimeException's, separate source, binary distributions.
>>
>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>> better. There are no real massive changes, we could continue in the
>> 0.10, 0.11, etc versions until the new api is done which would
>> become 1.0.0?

Re: 1.0.0 release?

["Mattmann, Chris A (388J)" <ch...@jpl.nasa.gov>]

I would also VOTE for 0.9.0 -- I find it confusing when projects don't follow an
incremental versioning approach...

Cheers,
Chris

On Mar 15, 2012, at 1:54 AM, Jarek Jarcec Cecho wrote:

> I'm not Brock nor committer, but I would like submit my feedback anyway :-)
> 
> Custom counter checking is still not in. I've submitted last version of my patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting for review. I'll be happy to finish the page into committable form.
> 
> I would personally vote for version 0.9.0, but I don't mind using 1.0.0 either.
> 
> Jarcec
> 
> On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
>> Brock,
>> 
>> What is your thought on doing a 1.0.0 release? I still would like to
>> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
>> maybe all new JIRA's could get scheduled for a different version,
>> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
>> mapreduce Combiner, custom counter checking, improved passing of the
>> conf, better support for java/other serializations, better error
>> messages, deprecated string methods, AssertionError's instead of
>> RuntimeException's, separate source, binary distributions.
>> 
>> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
>> better. There are no real massive changes, we could continue in the
>> 0.10, 0.11, etc versions until the new api is done which would
>> become 1.0.0?


++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Chris Mattmann, Ph.D.
Senior Computer Scientist
NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA
Office: 171-266B, Mailstop: 171-246
Email: chris.a.mattmann@nasa.gov
WWW:   http://sunset.usc.edu/~mattmann/
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Adjunct Assistant Professor, Computer Science Department
University of Southern California, Los Angeles, CA 90089 USA
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Re: 1.0.0 release?

[Jarek Jarcec Cecho <ja...@apache.org>]

I'm not Brock nor committer, but I would like submit my feedback anyway :-)

Custom counter checking is still not in. I've submitted last version of my patch to JIRA (add forgotten file CounterWrapper.java) and I'm waiting for review. I'll be happy to finish the page into committable form.

I would personally vote for version 0.9.0, but I don't mind using 1.0.0 either.

Jarcec

On Thu, Mar 15, 2012 at 04:47:58AM -0400, Jim Donofrio wrote:
> Brock,
> 
> What is your thought on doing a 1.0.0 release? I still would like to
> finish the 4 remaining unresolved JIRA's scheduled for 1.0.0 but
> maybe all new JIRA's could get scheduled for a different version,
> 1.0.1? or 1.1.0? I think we have enough new features/bug fixes:
> mapreduce Combiner, custom counter checking, improved passing of the
> conf, better support for java/other serializations, better error
> messages, deprecated string methods, AssertionError's instead of
> RuntimeException's, separate source, binary distributions.
> 
> Should the version really jump up to 1.0.0, maybe 0.9.0 would be
> better. There are no real massive changes, we could continue in the
> 0.10, 0.11, etc versions until the new api is done which would
> become 1.0.0?