Is there a "no graduate" option?

[Rich Bowen <rb...@rcbowen.com>]

Please forgive if this has been discussed in the past. I respond very 
well to "go look in the archives" responses. However, since folks' 
attention seems to be pointed this way, I thought I'd ask.

As I look through the list of projects in the incubator, trying to 
figure out where I can be useful, I notice several projects that have 
been in the incubator for a LONG time.

Which got me thinking ... is there a graceful exit option? I see on the 
list that one project is listed as a "Failed incubation", which seems 
really harsh. Is there an option for "You're very nice, but you just 
don't fit at the ASF"? Or is it assumed that the two options are 
"Graduate" and "Keep trying."

If a project has been in the incubator for more than 2 years, isn't it a 
safe assumption that they're not moving along very well, and we need to 
cut them loose and let them get on with their lives?

Perhaps, as long as we're talking of establishing limits, there needs to 
be a time deadline as well. Say, if you don't graduate in X TimeUnits 
(12 months? 24 months?) then we need to find a way to gracefully say, 
this isn't going to work out. It's a disservice to the project to keep 
them hanging if there's no real progress. It's a disservice to the 
infrastructure to keep them incubating if they're never going to hatch.

This isn't necessarily failure. There are a lot of good projects that 
just don't fit here. It's not a condemnation.

-- 
Rich Bowen
rbowen@rcbowen.com

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Henri Yandell <fl...@gmail.com>]

On 12/24/05, Sylvain Wallez <sy...@apache.org> wrote:
> Thomas Dudziak wrote:
>
> <snip/>
> > If now for instance, the DB PMC would somehow automatically get the
> > the incubation mails for the projects that it voted into incubation
> > and its reports, oversight from this PMC might enhance. After all it
> > involves more work to actively delete mails than to not get them in
> > the first place.
> >
> > And thus potential problems with incubated problems would surface
> > sooner, and the workload of the Incubator PMC might decrease - one
> > individual PMC has only a few projects in incubation whereas the
> > Incubator PMC has to take care of all incubated projects.
> >
>
> Interesting idea. Now forced distribution of emails won't work IMO as
> it's easy to delete without actually reading. So a way to ensure
> oversight of the originating PMC it to require a status update about
> related incubating projects in the quarterly board report.

+1. I'm operating under that already, though ApacheCon meant I screwed
up and didn't nudge Agila into providing a report.

Hen

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Rich Bowen <rb...@rcbowen.com>]

Sylvain Wallez wrote:
> Thomas Dudziak wrote:
> 
> <snip/>
>> If now for instance, the DB PMC would somehow automatically get the
>> the incubation mails for the projects that it voted into incubation
>> and its reports, oversight from this PMC might enhance. After all it
>> involves more work to actively delete mails than to not get them in
>> the first place.
>>
>> And thus potential problems with incubated problems would surface
>> sooner, and the workload of the Incubator PMC might decrease - one
>> individual PMC has only a few projects in incubation whereas the
>> Incubator PMC has to take care of all incubated projects.
>>   
> 
> Interesting idea. Now forced distribution of emails won't work IMO as 
> it's easy to delete without actually reading. So a way to ensure 
> oversight of the originating PMC it to require a status update about 
> related incubating projects in the quarterly board report.

+1. This is another "well, yes, *obviously* it should be that way" kind 
of idea. If the sponsoring PMC isn't taking an active role in the 
incubation process, it seems a pretty good chance that things are 
getting dropped or confused.

-- 
Rich Bowen
rbowen@rcbowen.com

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Sylvain Wallez <sy...@apache.org>]

Thomas Dudziak wrote:

<snip/>
> If now for instance, the DB PMC would somehow automatically get the
> the incubation mails for the projects that it voted into incubation
> and its reports, oversight from this PMC might enhance. After all it
> involves more work to actively delete mails than to not get them in
> the first place.
>
> And thus potential problems with incubated problems would surface
> sooner, and the workload of the Incubator PMC might decrease - one
> individual PMC has only a few projects in incubation whereas the
> Incubator PMC has to take care of all incubated projects.
>   

Interesting idea. Now forced distribution of emails won't work IMO as 
it's easy to delete without actually reading. So a way to ensure 
oversight of the originating PMC it to require a status update about 
related incubating projects in the quarterly board report.

Sylvain

-- 
Sylvain Wallez                        Anyware Technologies
http://bluxte.net                     http://www.anyware-tech.com
Apache Software Foundation Member     Research & Technology Director


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[robert burrell donkin <ro...@gmail.com>]

On 12/22/05, Thomas Dudziak <to...@gmail.com> wrote:

> I wonder whether there isn't another problem here: that of the
> (missing) oversight of the PMC that originally voted the project into
> incubation. For instance, while I'm on the DB PMC and voted e.g. Derby
> into incubation (just to pick an example), I did not really follow its
> way through incubation for all the usual reasons (too much work, own
> projects, bla, bla, bla). And I don't think that is much different for
> other PMCs.


+1

IMO (unlike most other pmc votes) ATM the way the process stands means that
there are very few negative consequences for a pmc (reputational or
otherwise) if a podling fails. this is one of the few areas at apache were
those who have to deal with any problems are not the ones with a vote. i'm
sure that some pmc's do care and this is the model that needs to be followed
but it's too easy at the moment for pmc's to vote yes without having to
worry about any negative consequences.

Now the problem that I see is that I was not 'forced' to take more
> care about the projects that I vote into incubation. Correct me if I'm
> wrong on this assumption, but in my naive view the incubator PMC has
> the role of ensuring that the project learns/follows the ASF ways,
> building a community etc. Whereas the DB PMC in my example would deal
> with things regarding the 'content' of the project.
>
> If now for instance, the DB PMC would somehow automatically get the
> the incubation mails for the projects that it voted into incubation
> and its reports, oversight from this PMC might enhance. After all it
> involves more work to actively delete mails than to not get them in
> the first place.
> And thus potential problems with incubated problems would surface
> sooner, and the workload of the Incubator PMC might decrease - one
> individual PMC has only a few projects in incubation whereas the
> Incubator PMC has to take care of all incubated projects.


not sure that all those emails would be very popular

but i agree that responsibility for oversight should be better aligned to
power. maybe it would just be a case of the pmc providing additional
oversight. this would mean that anyone voting +1 would be agreeing to
actively provide oversight (reading emails and so on).

perhaps pmc should be asked to include more details of the progress of their
sponsored podlings in the board reports.

- robert

Re: Is there a "no graduate" option?

[Thomas Dudziak <to...@gmail.com>]

> > It takes as long as it takes.
> >
> > The only requirement I have is that there is a continued effort towards
> > attracting community.  If a project goes completely dormant (i.e. no
> > traffic whatsoever), then yes it can fail and should be terminated.
> >
> > However, placing arbitrary limits on the Incubation period is the opposite
> > of what we want to achieve.  It would "force" a community to meet
> > artificial deadlines when that may be the worst thing we could do.
>
> +1
>
> but it would be a very good idea to have some sort of process for
> domancy: no development for even a month should ring alarm bells and
> call for either active intervention (if there is still a community) or
> a move to dormacy.

I wonder whether there isn't another problem here: that of the
(missing) oversight of the PMC that originally voted the project into
incubation. For instance, while I'm on the DB PMC and voted e.g. Derby
into incubation (just to pick an example), I did not really follow its
way through incubation for all the usual reasons (too much work, own
projects, bla, bla, bla). And I don't think that is much different for
other PMCs.

Now the problem that I see is that I was not 'forced' to take more
care about the projects that I vote into incubation. Correct me if I'm
wrong on this assumption, but in my naive view the incubator PMC has
the role of ensuring that the project learns/follows the ASF ways,
building a community etc. Whereas the DB PMC in my example would deal
with things regarding the 'content' of the project.

If now for instance, the DB PMC would somehow automatically get the
the incubation mails for the projects that it voted into incubation
and its reports, oversight from this PMC might enhance. After all it
involves more work to actively delete mails than to not get them in
the first place.
And thus potential problems with incubated problems would surface
sooner, and the workload of the Incubator PMC might decrease - one
individual PMC has only a few projects in incubation whereas the
Incubator PMC has to take care of all incubated projects.

Tom

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[robert burrell donkin <ro...@gmail.com>]

On 12/22/05, Justin Erenkrantz <ju...@erenkrantz.com> wrote:
> --On December 22, 2005 3:26:57 PM -0500 Rich Bowen <rb...@rcbowen.com>
> wrote:
>
> > As I look through the list of projects in the incubator, trying to figure
> > out where I can be useful, I notice several projects that have been in
> > the incubator for a LONG time.
> >
> > Which got me thinking ... is there a graceful exit option? I see on the
> > list that one project is listed as a "Failed incubation", which seems
> > really harsh. Is there an option for "You're very nice, but you just
> > don't fit at the ASF"?

interesting: i can't really see this happening for any project that
entered the incubator with the right spirit (it's tough being
rejected).they just tend to fade away.

but perhaps you were thinking of projects who entered with the wrong
spirit and refused to change. i'd hope that this would be an oversight
issue and would be dealt with appropriately by the incubator pmc.

> > Or is it assumed that the two options are
> > "Graduate" and "Keep trying."
>
> It takes as long as it takes.
>
> The only requirement I have is that there is a continued effort towards
> attracting community.  If a project goes completely dormant (i.e. no
> traffic whatsoever), then yes it can fail and should be terminated.
>
> However, placing arbitrary limits on the Incubation period is the opposite
> of what we want to achieve.  It would "force" a community to meet
> artificial deadlines when that may be the worst thing we could do.

+1

but it would be a very good idea to have some sort of process for
domancy: no development for even a month should ring alarm bells and
call for either active intervention (if there is still a community) or
a move to dormacy.

- robert

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

["Roy T. Fielding" <fi...@gbiv.com>]

On Dec 22, 2005, at 4:15 PM, Rich Bowen wrote:
> So, in your opinion, there is never a reason to tell a project, no,  
> you don't fit here? As long as they (think they) want to be at the  
> ASF, we should keep moving them in that direction?

That happens when a proposal gets no mentors.  It could also
happen if a podling disregards our policies and does something that
could harm the ASF.  The responsible PMC or the board can kill a
project any time it wants, for whatever reason, so there is no need
write down all the possible ways that a project can be terminated.

> Also, what signifies dormancy? Is there a time period after which  
> dormancy is indicated? Or is it the members of the project saying,  
> yes, we have abandoned the effort?

When the incubator PMC gets around to removing the podling, which
isn't going to happen until someone asks us to do so.  Usually it
is whoever is listed as mentoring the podling.

....Roy

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Rich Bowen <rb...@rcbowen.com>]

Justin Erenkrantz wrote:
> --On December 22, 2005 3:26:57 PM -0500 Rich Bowen <rb...@rcbowen.com> 
> wrote:
> 
>> As I look through the list of projects in the incubator, trying to figure
>> out where I can be useful, I notice several projects that have been in
>> the incubator for a LONG time.
>>
>> Which got me thinking ... is there a graceful exit option? I see on the
>> list that one project is listed as a "Failed incubation", which seems
>> really harsh. Is there an option for "You're very nice, but you just
>> don't fit at the ASF"? Or is it assumed that the two options are
>> "Graduate" and "Keep trying."
> 
> It takes as long as it takes.
> 
> The only requirement I have is that there is a continued effort towards 
> attracting community.  If a project goes completely dormant (i.e. no 
> traffic whatsoever), then yes it can fail and should be terminated.
> 
> However, placing arbitrary limits on the Incubation period is the 
> opposite of what we want to achieve.  It would "force" a community to 
> meet artificial deadlines when that may be the worst thing we could do.  

So, in your opinion, there is never a reason to tell a project, no, you 
don't fit here? As long as they (think they) want to be at the ASF, we 
should keep moving them in that direction?

Also, what signifies dormancy? Is there a time period after which 
dormancy is indicated? Or is it the members of the project saying, yes, 
we have abandoned the effort?

-- 
Rich Bowen
rbowen@rcbowen.com

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Justin Erenkrantz <ju...@erenkrantz.com>]

--On December 22, 2005 3:26:57 PM -0500 Rich Bowen <rb...@rcbowen.com> 
wrote:

> As I look through the list of projects in the incubator, trying to figure
> out where I can be useful, I notice several projects that have been in
> the incubator for a LONG time.
>
> Which got me thinking ... is there a graceful exit option? I see on the
> list that one project is listed as a "Failed incubation", which seems
> really harsh. Is there an option for "You're very nice, but you just
> don't fit at the ASF"? Or is it assumed that the two options are
> "Graduate" and "Keep trying."

It takes as long as it takes.

The only requirement I have is that there is a continued effort towards 
attracting community.  If a project goes completely dormant (i.e. no 
traffic whatsoever), then yes it can fail and should be terminated.

However, placing arbitrary limits on the Incubation period is the opposite 
of what we want to achieve.  It would "force" a community to meet 
artificial deadlines when that may be the worst thing we could do.  -- 
justin

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[J Aaron Farr <ja...@gmail.com>]

On 12/22/05, Rich Bowen <rb...@rcbowen.com> wrote:
> Davanum Srinivas wrote:
> > I'd say after X months (6? 12?), there should be a VOTE on incubator
> > PMC whether to continue that project or not.
>
> Ok, so say we went with X=12. There are 13 projects that fit that
> description. Of those, 3 have status updates that date within that X
> month timeframe, leaving 10 that, at least to me, appear to be inactive
> for more than X months.

I can speak a little for AltRMI and the FtpServer projects which were
spun out of Avalon oh so long ago.  This was early in the Incubator's
history before we had the guidelines and procedures we do today.  The
FtpServer has seen more recent activity, but those who have expressed
interested in AltRMI (myself included) simply haven't invested in
building a community around it (it was mostly stable code even when it
was in Avalon).  I believe a similar situation may be true of other
dormant incubator projects.

And "dormant" might be better than "failed."  If someone wants to come
along a pick up these projects and re-activate them, they are more
than welcome.

--
  jaaron

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Mads Toftum <ma...@toftum.dk>]

On Thu, Dec 22, 2005 at 04:51:42PM -0500, Rich Bowen wrote:
> Ok, so say we went with X=12. There are 13 projects that fit that 
> description. Of those, 3 have status updates that date within that X 
> month timeframe, leaving 10 that, at least to me, appear to be inactive 
> for more than X months.
> 
> So, say we were to vote on their continuation. Then what? Do we list 
> them as "failed"? That seems really harsh. Is there another term that 
> can be used?
> 
I'm not sure that "voting on their continuation" is the correct thing to
do, but I would certainly hope that the incubator pmc will take the
project up to revision at that point - probably even after 6 months of
no apparent activity. Auto terminating after a set period would be too
harsh, but reviewing and deciding wether to continue is reasonable.

vh

Mads Toftum
-- 
`Darn it, who spiked my coffee with water?!' - lwall


---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Rich Bowen <rb...@rcbowen.com>]

Davanum Srinivas wrote:
> I'd say after X months (6? 12?), there should be a VOTE on incubator
> PMC whether to continue that project or not.

Ok, so say we went with X=12. There are 13 projects that fit that 
description. Of those, 3 have status updates that date within that X 
month timeframe, leaving 10 that, at least to me, appear to be inactive 
for more than X months.

So, say we were to vote on their continuation. Then what? Do we list 
them as "failed"? That seems really harsh. Is there another term that 
can be used?

-- 
Rich Bowen
rbowen@rcbowen.com

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org

Re: Is there a "no graduate" option?

[Davanum Srinivas <da...@gmail.com>]

I'd say after X months (6? 12?), there should be a VOTE on incubator
PMC whether to continue that project or not.

thanks,
dims

On 12/22/05, Rich Bowen <rb...@rcbowen.com> wrote:
> Please forgive if this has been discussed in the past. I respond very
> well to "go look in the archives" responses. However, since folks'
> attention seems to be pointed this way, I thought I'd ask.
>
> As I look through the list of projects in the incubator, trying to
> figure out where I can be useful, I notice several projects that have
> been in the incubator for a LONG time.
>
> Which got me thinking ... is there a graceful exit option? I see on the
> list that one project is listed as a "Failed incubation", which seems
> really harsh. Is there an option for "You're very nice, but you just
> don't fit at the ASF"? Or is it assumed that the two options are
> "Graduate" and "Keep trying."
>
> If a project has been in the incubator for more than 2 years, isn't it a
> safe assumption that they're not moving along very well, and we need to
> cut them loose and let them get on with their lives?
>
> Perhaps, as long as we're talking of establishing limits, there needs to
> be a time deadline as well. Say, if you don't graduate in X TimeUnits
> (12 months? 24 months?) then we need to find a way to gracefully say,
> this isn't going to work out. It's a disservice to the project to keep
> them hanging if there's no real progress. It's a disservice to the
> infrastructure to keep them incubating if they're never going to hatch.
>
> This isn't necessarily failure. There are a lot of good projects that
> just don't fit here. It's not a condemnation.
>
> --
> Rich Bowen
> rbowen@rcbowen.com
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
>


--
Davanum Srinivas : http://wso2.com/blogs/

---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org