You are viewing a plain text version of this content. The canonical link for it is here.
Posted to labs@labs.apache.org by Danny Angus <da...@apache.org> on 2017/03/08 15:27:56 UTC
Releases
Hi
I would love to be the kind of guy who would set out a series of topics
that we need to discuss, contextualise them and initiate each one,
collating the output at the end, but I'm not.
So I have a question for the people who have mentioned lack of releases as
a problem..
The idea behind "no releases" was that a release implies that certain
standards common to the ASF are being met, standards which Labs don't
apply, and that if a lab needs to cut a release it probably has enough
users and maintainers to become a project of its own.
So my question to those people is this; Is it the lack of the formality and
assurance that is a problem or the lack of a build? Would "pre-release
downloads" be enough?
d.
Re: Releases
Posted by Greg Brown <gk...@icloud.com>.
> ….even a one person community want to have “something” out there, that is equal to a release. So no the term “pre-release downloads” would not really make many happy (that can be done and are done already today).
...
> If you compare with GitHub, I can make a release every day if I want (and call it a release), so why limit myself ?
+1
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Greg Brown <gk...@icloud.com>.
>> Never got beyond the stage of a one man show.
>
> Did you know the incubator was the next step in the process?
I also manage a couple of "one man show" projects, and was told that the Incubator was not the right place for them. Given the ASF's emphasis on community, it would be great if we could provide a place to foster such projects and help them develop into something that is actually suitable for the Incubator.
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Graham Leggett <mi...@sharp.fm>.
On 11 Mar 2017, at 2:16 PM, Jochen Wiedmann <jo...@gmail.com> wrote:
>> Did you try and enter the incubator?
>
> Never got beyond the stage of a one man show.
Did you know the incubator was the next step in the process?
Regards,
Graham
—
Re: Releases
Posted by jan iversen <ja...@gmail.com>.
> On 14 Mar 2017, at 08:05, Jochen Wiedmann <jo...@gmail.com> wrote:
>
> On Mon, Mar 13, 2017 at 6:32 PM, Danny Angus <da...@apache.org> wrote:
>> So what are the key characteristics of a "release"?
>> A version number?
>> A permalinked immutable artifact?
>
> Definitely these. Ideally, the additional ability to distribute where
> developers are looking for such stuff (aka repo.maven.org).
>
> In essence, a reference, which others may use when starting to use
> that stuff in external projects. (As opposed to instructions like
> "Build it”.)
A reference (a version) is important, as well as a place to download it from, not only as source, but frequently also as jar/library/etc.
>
>> A test completion certificate? Signed distributables?
>
> I am unaware of someone really bothering about either of these.
Signing does not seem to be a major problem, unless you work on something for the apple store, and can in that case easily be done by one of the developers.
However having some short of SHA checksum is important especially for prebuilt libraries
Whether it is called release, release-LABS or a third option, is not a problem, it is the characteristics that matter.
rgds
jan I.
>
> Jochen
>
>
> --
> The next time you hear: "Don't reinvent the wheel!"
>
> http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/evolution-of-the-wheel-300x85.jpg
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
> For additional commands, e-mail: labs-help@labs.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Jochen Wiedmann <jo...@gmail.com>.
On Mon, Mar 13, 2017 at 6:32 PM, Danny Angus <da...@apache.org> wrote:
> So what are the key characteristics of a "release"?
> A version number?
> A permalinked immutable artifact?
Definitely these. Ideally, the additional ability to distribute where
developers are looking for such stuff (aka repo.maven.org).
In essence, a reference, which others may use when starting to use
that stuff in external projects. (As opposed to instructions like
"Build it".)
> A test completion certificate? Signed distributables?
I am unaware of someone really bothering about either of these.
Jochen
--
The next time you hear: "Don't reinvent the wheel!"
http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/evolution-of-the-wheel-300x85.jpg
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Danny Angus <da...@apache.org>.
So what are the key characteristics of a "release"?
A version number?
A permalinked immutable artifact?
A test completion certificate? Signed distributables?
What was it that would have made a difference to you?
d.
On 11 March 2017 at 12:16, Jochen Wiedmann <jo...@gmail.com> wrote:
> On Fri, Mar 10, 2017 at 3:29 PM, Graham Leggett <mi...@sharp.fm> wrote:
>
>> Did you try and enter the incubator?
>
> Never got beyond the stage of a one man show.
>
> Jochen
>
>
> --
> The next time you hear: "Don't reinvent the wheel!"
>
> http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/evolution-of-the-wheel-300x85.jpg
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
> For additional commands, e-mail: labs-help@labs.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Jochen Wiedmann <jo...@gmail.com>.
On Fri, Mar 10, 2017 at 3:29 PM, Graham Leggett <mi...@sharp.fm> wrote:
> Did you try and enter the incubator?
Never got beyond the stage of a one man show.
Jochen
--
The next time you hear: "Don't reinvent the wheel!"
http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/evolution-of-the-wheel-300x85.jpg
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Graham Leggett <mi...@sharp.fm>.
On 10 Mar 2017, at 3:49 PM, Jochen Wiedmann <jo...@gmail.com> wrote:
>> I think the purpose of Labs should be this:
>>
>> “Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
>>
>> The point at which you enter the Apache Incubator is the point at which you want to make your first release.
>
> Meaning, you wish things to stay exactly as they are.
>
> To me, the above has been the precise reason, why Labs has failed (for
> me), and why I left it.
Did you try and enter the incubator?
Regards,
Graham
—
Re: Releases
Posted by Jochen Wiedmann <jo...@gmail.com>.
On Fri, Mar 10, 2017 at 12:45 PM, Graham Leggett <mi...@sharp.fm> wrote:
> I think the purpose of Labs should be this:
>
> “Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
>
> The point at which you enter the Apache Incubator is the point at which you want to make your first release.
Meaning, you wish things to stay exactly as they are.
To me, the above has been the precise reason, why Labs has failed (for
me), and why I left it.
Jochen
--
The next time you hear: "Don't reinvent the wheel!"
http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/evolution-of-the-wheel-300x85.jpg
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Danny Angus <da...@apache.org>.
+1
On 10 March 2017 at 11:45, Graham Leggett <mi...@sharp.fm> wrote:
> On 08 Mar 2017, at 7:15 PM, jan iversen <ja...@gmail.com> wrote:
>
>> If you compare with GitHub, I can make a release every day if I want (and call it a release), so why limit myself ?
>
> You can release every day at Github sure, but you can’t call those releases Apache releases. To be released by the Apache Software Foundation means something very specific.
>
> I think the purpose of Labs should be this:
>
> “Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
>
> The point at which you enter the Apache Incubator is the point at which you want to make your first release.
>
> This discussion should really happen in collaboration with the incubator folks. Ideally Labs and the Incubator should work closely together, or the Labs project should even be a subproject of the incubator.
>
> Regards,
> Graham
> —
>
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Raphael Bircher <rb...@gmail.com>.
Am .03.2017, 13:22 Uhr, schrieb Greg Brown <gk...@icloud.com>:
>> I think the purpose of Labs should be this:
>>
>> “Labs is the place where you develop software with the goal of
>> introducing that software to the Apache Incubator”.
>>
>> The point at which you enter the Apache Incubator is the point at which
>> you want to make your first release.
>
> +1
+1 for me too
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
> For additional commands, e-mail: labs-help@labs.apache.org
>
--
My introduction https://youtu.be/Ln4vly5sxYU
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Greg Brown <gk...@icloud.com>.
> I think the purpose of Labs should be this:
>
> “Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
>
> The point at which you enter the Apache Incubator is the point at which you want to make your first release.
+1
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Wade Chandler <wa...@apache.org>.
> On Mar 10, 2017, at 10:21, Raphael Bircher <rb...@gmail.com> wrote:
>
> Hi Graham
>
> Am .03.2017, 15:34 Uhr, schrieb Graham Leggett <minfrin@sharp.fm <ma...@sharp.fm>>:
>
>> On 10 Mar 2017, at 3:48 PM, Hadrian Zbarcea <hz...@gmail.com> wrote:
>>
>>>> I think the purpose of Labs should be this:
>>>>
>>>> “Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
>>> So you *really* believe that an labs being a incubator-incubator or pre-incubator will provide enough value to attract committers to start here instead of github?
>>
>> It depends on whether people want to be an Apache project or not.
>>
>>> Isn't it exactly what labs are about today? Labs didn't attract many projects in the past years, what makes us think that they will in the future?
>>
>> If Labs positions itself properly and explains how it fits into the overall Apache picture, I don’t see why this wouldn’t be a problem.
>>
>> When people demand releases out of Labs, it means they don’t yet understand how Apache works, or what Apache is supposed to do for them. In cases like this is the proposed project a fit for the ASF?
> At the moment it looks for me as you are against any change. But you don't tell us any reason for it, You just say no. Is this Apache Way? Not really, right?
>
> No one ask to do a full release out of labs. We talked about snapshots or something similar. And BTW most Project who enters the incubator have had already a release outside incubator.
>
> I wonder what are your arguments against a snapshot (or whatever we call this) out of the labs.
>
This conversation and call for action started off by first stating that labs wasn’t really a success, and asked for some input, as well as if anyone was interested in shaping it’s future. So, I propose some focus be given to the question and the reasonings for the call to action.
It seems fairly clear one reason is it wasn’t exactly easily discovered nor prominently mentioned where new comers would have seen it nor come to understand it. But, there are surely other reasons, and ones besides it just wasn’t documented well enough to fit into “The Apache Way”. I think when people like myself first read the welcome section at http://labs.apache.org <http://labs.apache.org/> there are some very specific things they will pick out, such as “Apache Labs is a place for innovation” and “without discrimination of purpose, medium, or implementation technology” and “provide the necessary resource to promote and maintain the innovative power within the Apache community”.
I think this discussion should perhaps first focus on what is good about that introduction, and ask what is appealing about labs to those who had never heard of it, or those who left it plus their reasonings for doing that. Releases or some notion of how to more easily use labs projects for other types of innovation, such as Maven projects, is probably a good way to reduce friction.
Not everything in a playground should become a fit for the organization for which it was an experiment or exercise in creativity or innovation. But, that shouldn’t mean one doesn’t want some people to try it out before that is decided. Given X libraries in some experimental state, perhaps a lab only makes sense because of the others; if they didn’t exist first, could there have been innovation or vision to use them together in a particular way? This notion of a piece of software as a binary artifact making it easier to link with others being exactly the same thing as a release should really be dwelled upon. That simply isn’t true in very many places and situations.
In my opinion, if labs has too much friction, then it can’t be called a play ground or a place for innovation. If it is where things start out fresh before they go into the incubator, if they are started at ASF, then that could still happen as part of a bigger process, but if that is the only reason for labs, then that is the only time it will be used. If the goal is to give committers a little room to play and experiment, then it isn’t going to always be that they think that item needs to be an ASF project nor brought to the incubator; they really would just be doing that experimentation as a group of fellow ASF members, and that is the reason to do such things here versus some where else outside of legal protections the collective and policies bring.
If there are specifics on what can not be done per very specific legal reasons that is one thing. Those should be the things mentioned as the limits, and clearly stated as to why. To say “that isn’t the ASF way, and someone doesn’t know that yet” doesn’t necessarily work to solve why labs wasn’t successful nor what it’s future could be.
Thanks,
Wade
Re: Releases
Posted by Graham Leggett <mi...@sharp.fm>.
On 10 Mar 2017, at 5:21 PM, Raphael Bircher <rb...@gmail.com> wrote:
> At the moment it looks for me as you are against any change. But you don't tell us any reason for it, You just say no. Is this Apache Way? Not really, right?
That’s too general a statement to respond to.
> No one ask to do a full release out of labs. We talked about snapshots or something similar. And BTW most Project who enters the incubator have had already a release outside incubator.
>
> I wonder what are your arguments against a snapshot (or whatever we call this) out of the labs.
The title of this thread is “Releases”.
Regards,
Graham
—
Re: Releases
Posted by Raphael Bircher <rb...@gmail.com>.
Hi Graham
Am .03.2017, 15:34 Uhr, schrieb Graham Leggett <mi...@sharp.fm>:
> On 10 Mar 2017, at 3:48 PM, Hadrian Zbarcea <hz...@gmail.com> wrote:
>
>>> I think the purpose of Labs should be this:
>>>
>>> “Labs is the place where you develop software with the goal of
>>> introducing that software to the Apache Incubator”.
>> So you *really* believe that an labs being a incubator-incubator or
>> pre-incubator will provide enough value to attract committers to start
>> here instead of github?
>
> It depends on whether people want to be an Apache project or not.
>
>> Isn't it exactly what labs are about today? Labs didn't attract many
>> projects in the past years, what makes us think that they will in the
>> future?
>
> If Labs positions itself properly and explains how it fits into the
> overall Apache picture, I don’t see why this wouldn’t be a problem.
>
> When people demand releases out of Labs, it means they don’t yet
> understand how Apache works, or what Apache is supposed to do for them.
> In cases like this is the proposed project a fit for the ASF?
At the moment it looks for me as you are against any change. But you don't
tell us any reason for it, You just say no. Is this Apache Way? Not
really, right?
No one ask to do a full release out of labs. We talked about snapshots or
something similar. And BTW most Project who enters the incubator have had
already a release outside incubator.
I wonder what are your arguments against a snapshot (or whatever we call
this) out of the labs.
Regards, Raphael
--
My introduction https://youtu.be/Ln4vly5sxYU
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Graham Leggett <mi...@sharp.fm>.
On 10 Mar 2017, at 3:48 PM, Hadrian Zbarcea <hz...@gmail.com> wrote:
>> I think the purpose of Labs should be this:
>>
>> “Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
> So you *really* believe that an labs being a incubator-incubator or pre-incubator will provide enough value to attract committers to start here instead of github?
It depends on whether people want to be an Apache project or not.
> Isn't it exactly what labs are about today? Labs didn't attract many projects in the past years, what makes us think that they will in the future?
If Labs positions itself properly and explains how it fits into the overall Apache picture, I don’t see why this wouldn’t be a problem.
When people demand releases out of Labs, it means they don’t yet understand how Apache works, or what Apache is supposed to do for them. In cases like this is the proposed project a fit for the ASF?
>> The point at which you enter the Apache Incubator is the point at which you want to make your first release.
>>
>> This discussion should really happen in collaboration with the incubator folks. Ideally Labs and the Incubator should work closely together, or the Labs project should even be a subproject of the incubator.
> My understanding is that projects are meant to graduate out of the incubator. This proposal implies that labs@ are not meant to graduate, right?
Not following the logic, can you explain again?
Regards,
Graham
—
Re: Releases
Posted by Hadrian Zbarcea <hz...@gmail.com>.
On 03/10/2017 06:45 AM, Graham Leggett wrote:
> On 08 Mar 2017, at 7:15 PM, jan iversen <ja...@gmail.com> wrote:
>
>> If you compare with GitHub, I can make a release every day if I want (and call it a release), so why limit myself ?
>
> You can release every day at Github sure, but you can\u2019t call those releases Apache releases. To be released by the Apache Software Foundation means something very specific.
>
> I think the purpose of Labs should be this:
>
> \u201cLabs is the place where you develop software with the goal of introducing that software to the Apache Incubator\u201d.
So you *really* believe that an labs being a incubator-incubator or
pre-incubator will provide enough value to attract committers to start
here instead of github?
Isn't it exactly what labs are about today? Labs didn't attract many
projects in the past years, what makes us think that they will in the
future?
>
> The point at which you enter the Apache Incubator is the point at which you want to make your first release.
>
> This discussion should really happen in collaboration with the incubator folks. Ideally Labs and the Incubator should work closely together, or the Labs project should even be a subproject of the incubator.
My understanding is that projects are meant to graduate out of the
incubator. This proposal implies that labs@ are not meant to graduate,
right?
>
> Regards,
> Graham
> \u2014
>
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Atri Sharma <at...@apache.org>.
+1
On Mar 10, 2017 5:15 PM, "Graham Leggett" <mi...@sharp.fm> wrote:
> On 08 Mar 2017, at 7:15 PM, jan iversen <ja...@gmail.com> wrote:
>
> > If you compare with GitHub, I can make a release every day if I want
> (and call it a release), so why limit myself ?
>
> You can release every day at Github sure, but you can’t call those
> releases Apache releases. To be released by the Apache Software Foundation
> means something very specific.
>
> I think the purpose of Labs should be this:
>
> “Labs is the place where you develop software with the goal of introducing
> that software to the Apache Incubator”.
>
> The point at which you enter the Apache Incubator is the point at which
> you want to make your first release.
>
> This discussion should really happen in collaboration with the incubator
> folks. Ideally Labs and the Incubator should work closely together, or the
> Labs project should even be a subproject of the incubator.
>
> Regards,
> Graham
> —
>
>
Re: Releases
Posted by Graham Leggett <mi...@sharp.fm>.
On 08 Mar 2017, at 7:15 PM, jan iversen <ja...@gmail.com> wrote:
> If you compare with GitHub, I can make a release every day if I want (and call it a release), so why limit myself ?
You can release every day at Github sure, but you can’t call those releases Apache releases. To be released by the Apache Software Foundation means something very specific.
I think the purpose of Labs should be this:
“Labs is the place where you develop software with the goal of introducing that software to the Apache Incubator”.
The point at which you enter the Apache Incubator is the point at which you want to make your first release.
This discussion should really happen in collaboration with the incubator folks. Ideally Labs and the Incubator should work closely together, or the Labs project should even be a subproject of the incubator.
Regards,
Graham
—
Re: Releases
Posted by jan iversen <ja...@gmail.com>.
> On 8 Mar 2017, at 16:27, Danny Angus <da...@apache.org> wrote:
>
> Hi
> I would love to be the kind of guy who would set out a series of topics
> that we need to discuss, contextualise them and initiate each one,
> collating the output at the end, but I'm not.
>
> So I have a question for the people who have mentioned lack of releases as
> a problem..
>
> The idea behind "no releases" was that a release implies that certain
> standards common to the ASF are being met, standards which Labs don't
> apply, and that if a lab needs to cut a release it probably has enough
> users and maintainers to become a project of its own.
>
> So my question to those people is this; Is it the lack of the formality and
> assurance that is a problem or the lack of a build? Would "pre-release
> downloads" be enough?
Seen from what I learned….even a one person community want to have “something” out there, that is equal to a release. So no the term “pre-release downloads” would not really make many happy (that can be done and are done already today).
If you look at incubator, they have a slightly different set of release rules. I had a vision of something called “LABS-Release, limited responsibility”. We would still do the IP check etc, but not have a 3 person PMC voting.
If you compare with GitHub, I can make a release every day if I want (and call it a release), so why limit myself ?
rgds
jan I.
>
> d.
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org
Re: Releases
Posted by Dennis Gove <dp...@gmail.com>.
I think yes, but limited to
https://repository.apache.org/content/repositories/snapshots, only.
On Fri, Mar 10, 2017 at 6:08 AM, Jochen Wiedmann <jo...@gmail.com>
wrote:
> On Wed, Mar 8, 2017 at 4:27 PM, Danny Angus <da...@apache.org> wrote:
>
> > So my question to those people is this; Is it the lack of the formality
> and
> > assurance that is a problem or the lack of a build? Would "pre-release
> > downloads" be enough?
>
> Would have been enough for me. OTOH: With such "pre-releases", the
> first question, that comes to mind: Publishable on repo.maven.org?
>
> Jochen
>
>
> --
> The next time you hear: "Don't reinvent the wheel!"
>
> http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/
> evolution-of-the-wheel-300x85.jpg
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
> For additional commands, e-mail: labs-help@labs.apache.org
>
>
Re: Releases
Posted by Jochen Wiedmann <jo...@gmail.com>.
On Wed, Mar 8, 2017 at 4:27 PM, Danny Angus <da...@apache.org> wrote:
> So my question to those people is this; Is it the lack of the formality and
> assurance that is a problem or the lack of a build? Would "pre-release
> downloads" be enough?
Would have been enough for me. OTOH: With such "pre-releases", the
first question, that comes to mind: Publishable on repo.maven.org?
Jochen
--
The next time you hear: "Don't reinvent the wheel!"
http://www.keystonedevelopment.co.uk/wp-content/uploads/2014/10/evolution-of-the-wheel-300x85.jpg
---------------------------------------------------------------------
To unsubscribe, e-mail: labs-unsubscribe@labs.apache.org
For additional commands, e-mail: labs-help@labs.apache.org