Re: ECharts incubator report
As a mentor, I've been following this. There is a legal concern here so it is best if we get on the right track as soon as possible. I understand removing things from npm is a PITA. But retagging is easy. The biggest issue seems to be retagging the rc1. Can you get an rc2 or ideally a proper release voted on and up on npm ASAP and then tag rc1 as deprecated? Is there concern about using the deprecated tag? Even if rc2 is just a more formally voted release with a version change, getting it up there and tagged with rc seems to be a good fix. I would recommend not going to the board. That will just be discussed and likely kicked back for more discussion causing you months of delay. Regards, KAM - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Using "rc" tag for release candidates seems to be a reasonable choice, in which case developers need to explicitly use "npm install echarts@rc" to use the rc version. I think Justin has no objection to this, right? But our main controversy lies on what to do with v4.2.0-rc1, which has already been published to "latest" and can be accessed by "npm install". The reasons were stated clearly in previous emails. We would like to get some advice from mentors that whether there could be some way to prevent the mess caused by unpublishing an rc version without a new correct version, and whether it is necessary for us to ask ASF board about it. Zhang Wenli http://zhangwenli.com On Sun, Feb 3, 2019 at 1:40 PM Dave Fisher wrote: > Please do not use Cordova as an example of a conforming project. > > Let’s discuss how to properly use of npm and staying where snapshots > require explicit user choice. > > Regards, > Dave > > Sent from my iPhone > > > On Feb 2, 2019, at 3:52 PM, Clement Ho > wrote: > > > > I just want to note that another popular apache project, cordova seems to > > publish nightly builds to npm ( > > https://www.npmjs.com/package/cordova) > > > > But it does look like the ‘latest’ tag is always set to the approved > > release for public use. Perhaps echarts can mimic their exact model? > > > > Clement > > > >> On Sun, Feb 3, 2019 at 7:28 AM Justin Mclean > wrote: > >> > >> Hi, > >> > >> I can see that nothing has been done about [1]. The PPMC cannot provide > >> unapproved releases to the general public in this manner. [2] All > releases > >> must be approved by the PPMC and IPMC while in incubation. As the VP of > the > >> Apache Incubator I'm asking you to please remove this unapproved release > >> and any others you have made public since starting graduation. > >> > >> If you can come up with another solution that complies with the ASF > >> release policy I'm willing to listen to it. > >> > >> Thanks, > >> Justin > >> V.P. Incubator > >> > >> 1. https://www.npmjs.com/package/echarts > >> 2. http://www.apache.org/legal/release-policy.html#release-approval > >> > >> - > >> To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > >> For additional commands, e-mail: dev-h...@echarts.apache.org > >> > >> > > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > >
Re: ECharts incubator report
Please do not use Cordova as an example of a conforming project. Let’s discuss how to properly use of npm and staying where snapshots require explicit user choice. Regards, Dave Sent from my iPhone > On Feb 2, 2019, at 3:52 PM, Clement Ho wrote: > > I just want to note that another popular apache project, cordova seems to > publish nightly builds to npm ( > https://www.npmjs.com/package/cordova) > > But it does look like the ‘latest’ tag is always set to the approved > release for public use. Perhaps echarts can mimic their exact model? > > Clement > >> On Sun, Feb 3, 2019 at 7:28 AM Justin Mclean wrote: >> >> Hi, >> >> I can see that nothing has been done about [1]. The PPMC cannot provide >> unapproved releases to the general public in this manner. [2] All releases >> must be approved by the PPMC and IPMC while in incubation. As the VP of the >> Apache Incubator I'm asking you to please remove this unapproved release >> and any others you have made public since starting graduation. >> >> If you can come up with another solution that complies with the ASF >> release policy I'm willing to listen to it. >> >> Thanks, >> Justin >> V.P. Incubator >> >> 1. https://www.npmjs.com/package/echarts >> 2. http://www.apache.org/legal/release-policy.html#release-approval >> >> - >> To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org >> For additional commands, e-mail: dev-h...@echarts.apache.org >> >> - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Thanks Justin. I think we should probably ask the ASF board about this. In the meantime, could our mentors give some advice and opinions on this? Zhang Wenli http://zhangwenli.com On Sun, Feb 3, 2019 at 12:32 PM Justin Mclean wrote: > Hi, > > I sorry but what npm recommends is nothing to do with ASF release policy. > > > On the other hand, if a version is deprecated, it will give a warning to > tell > > the developers to update, which, in our case, is somewhat strange since > we > > don't have a newer correct version. > > Yes that is unfortunate but you need to follow ASF release policy. If you > want an exception to this then you need to ask the ASF board. > > > So, the safest and best-for-all solution seems to be, releasing a correct > > version with the permission of PPMC ASAP, which we are working on > > currently. And after that, mark the 4.2.0-rc1 to be deprecated. > > I don't think that is the best course of action but discuss with your > mentors and see what they say. > > > I think this way also benefits Apache's branding in saving a large number > > of developers' confusion and complaints. > > Could @Justin please help judge if this works with Apache? > > By putting out unapproved released you put the legal protections that the > ASF have in place for your project and the ASF itself at risk. It's > probably not a large risk but it's there all the same. Please read [1] so > you understand why the ASF has a release policy and why we don't allow > unapproved releases. > > Thanks, > Justin > > 1. http://www.apache.org/legal/release-policy.html#why > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > >
Re: ECharts incubator report
In my prior experience working with unpublishing npm packages, the npm support team will only let you deprecate packages when you are within a certain download volume - which I think echarts would fit in that category. At GitLab (where I work), we had to deprecate vs unpublishing and we had a lower weekly download count than echarts. On Sun, Feb 3, 2019 at 12:32 PM Justin Mclean wrote: > Hi, > > I sorry but what npm recommends is nothing to do with ASF release policy. > > > On the other hand, if a version is deprecated, it will give a warning to > tell > > the developers to update, which, in our case, is somewhat strange since > we > > don't have a newer correct version. > > Yes that is unfortunate but you need to follow ASF release policy. If you > want an exception to this then you need to ask the ASF board. > > > So, the safest and best-for-all solution seems to be, releasing a correct > > version with the permission of PPMC ASAP, which we are working on > > currently. And after that, mark the 4.2.0-rc1 to be deprecated. > > I don't think that is the best course of action but discuss with your > mentors and see what they say. > > > I think this way also benefits Apache's branding in saving a large number > > of developers' confusion and complaints. > > Could @Justin please help judge if this works with Apache? > > By putting out unapproved released you put the legal protections that the > ASF have in place for your project and the ASF itself at risk. It's > probably not a large risk but it's there all the same. Please read [1] so > you understand why the ASF has a release policy and why we don't allow > unapproved releases. > > Thanks, > Justin > > 1. http://www.apache.org/legal/release-policy.html#why > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > >
Re: ECharts incubator report
Hi, I sorry but what npm recommends is nothing to do with ASF release policy. > On the other hand, if a version is deprecated, it will give a warning to tell > the developers to update, which, in our case, is somewhat strange since we > don't have a newer correct version. Yes that is unfortunate but you need to follow ASF release policy. If you want an exception to this then you need to ask the ASF board. > So, the safest and best-for-all solution seems to be, releasing a correct > version with the permission of PPMC ASAP, which we are working on > currently. And after that, mark the 4.2.0-rc1 to be deprecated. I don't think that is the best course of action but discuss with your mentors and see what they say. > I think this way also benefits Apache's branding in saving a large number > of developers' confusion and complaints. > Could @Justin please help judge if this works with Apache? By putting out unapproved released you put the legal protections that the ASF have in place for your project and the ASF itself at risk. It's probably not a large risk but it's there all the same. Please read [1] so you understand why the ASF has a release policy and why we don't allow unapproved releases. Thanks, Justin 1. http://www.apache.org/legal/release-policy.html#why - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Thanks Clement. Yes, I think we and Justin have reached such a consensus that we should use npm tag for rc releases. What we are discussing about is what we should do to the previous version 4.2.0-rc1, which was published without the rc tag since we didn't know the method yet. Zhang Wenli http://zhangwenli.com On Sun, Feb 3, 2019 at 7:52 AM Clement Ho wrote: > I just want to note that another popular apache project, cordova seems to > publish nightly builds to npm ( > https://www.npmjs.com/package/cordova) > > But it does look like the ‘latest’ tag is always set to the approved > release for public use. Perhaps echarts can mimic their exact model? > > Clement > > On Sun, Feb 3, 2019 at 7:28 AM Justin Mclean wrote: > > > Hi, > > > > I can see that nothing has been done about [1]. The PPMC cannot provide > > unapproved releases to the general public in this manner. [2] All > releases > > must be approved by the PPMC and IPMC while in incubation. As the VP of > the > > Apache Incubator I'm asking you to please remove this unapproved release > > and any others you have made public since starting graduation. > > > > If you can come up with another solution that complies with the ASF > > release policy I'm willing to listen to it. > > > > Thanks, > > Justin > > V.P. Incubator > > > > 1. https://www.npmjs.com/package/echarts > > 2. http://www.apache.org/legal/release-policy.html#release-approval > > > > - > > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > > For additional commands, e-mail: dev-h...@echarts.apache.org > > > > >
Re: ECharts incubator report
Hi Justin, We are willing to deal with the latest released rc version problem. As we can see from npm documentation [1], unpublishing is "generally considered bad behavior to remove versions of a library that others are depending on". And in our case, it is even more dangerous since ECharts has more than 70,000 downloads, and we haven't provided a correct version after that. It's recommended to mark the previous rc version to be deprecated rather than unpublish it [1][2]. And please note that > With the default registry (registry.npmjs.org), unpublish is only allowed > with versions published in the last 72 hours. If you are trying to > unpublish a version published longer ago than that, contact > supp...@npmjs.com. [1] When a version is unpublished, if some projects have something like `"echarts": "^4.2.0-rc1"` in their package.json, it will cause an error in `npm install` step and bring confusion to a great number of developers. On the other hand, if a version is deprecated, it will give a warning to tell the developers to update, which, in our case, is somewhat strange since we don't have a newer correct version. So, the safest and best-for-all solution seems to be, releasing a correct version with the permission of PPMC ASAP, which we are working on currently. And after that, mark the 4.2.0-rc1 to be deprecated. And we promise that all rc releases under npm in the future should go with "rc" tag to prevent it from being "latest", and the default install way "npm install" will use the latest version permitted by PPMC. I think this way also benefits Apache's branding in saving a large number of developers' confusion and complaints. Could @Justin please help judge if this works with Apache? BTW, we are going to have Chinese New Year from tomorrow till next week, so slow email response is expected. Sorry for the inconvenience, and wish you a happy new year! [1] https://docs.npmjs.com/cli/unpublish [2] https://docs.npmjs.com/cli/deprecate Zhang Wenli http://zhangwenli.com On Sun, Feb 3, 2019 at 7:28 AM Justin Mclean wrote: > Hi, > > I can see that nothing has been done about [1]. The PPMC cannot provide > unapproved releases to the general public in this manner. [2] All releases > must be approved by the PPMC and IPMC while in incubation. As the VP of the > Apache Incubator I'm asking you to please remove this unapproved release > and any others you have made public since starting graduation. > > If you can come up with another solution that complies with the ASF > release policy I'm willing to listen to it. > > Thanks, > Justin > V.P. Incubator > > 1. https://www.npmjs.com/package/echarts > 2. http://www.apache.org/legal/release-policy.html#release-approval > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > >
Re: ECharts incubator report
I just want to note that another popular apache project, cordova seems to publish nightly builds to npm ( https://www.npmjs.com/package/cordova) But it does look like the ‘latest’ tag is always set to the approved release for public use. Perhaps echarts can mimic their exact model? Clement On Sun, Feb 3, 2019 at 7:28 AM Justin Mclean wrote: > Hi, > > I can see that nothing has been done about [1]. The PPMC cannot provide > unapproved releases to the general public in this manner. [2] All releases > must be approved by the PPMC and IPMC while in incubation. As the VP of the > Apache Incubator I'm asking you to please remove this unapproved release > and any others you have made public since starting graduation. > > If you can come up with another solution that complies with the ASF > release policy I'm willing to listen to it. > > Thanks, > Justin > V.P. Incubator > > 1. https://www.npmjs.com/package/echarts > 2. http://www.apache.org/legal/release-policy.html#release-approval > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > >
Re: ECharts incubator report
Hi, I can see that nothing has been done about [1]. The PPMC cannot provide unapproved releases to the general public in this manner. [2] All releases must be approved by the PPMC and IPMC while in incubation. As the VP of the Apache Incubator I'm asking you to please remove this unapproved release and any others you have made public since starting graduation. If you can come up with another solution that complies with the ASF release policy I'm willing to listen to it. Thanks, Justin V.P. Incubator 1. https://www.npmjs.com/package/echarts 2. http://www.apache.org/legal/release-policy.html#release-approval - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Hi, > But the rc version is not tagged as "latest" in npm, so the public can not > get this version by `npm install`. That seems acceptable / reasonable thing to do as only people who know it exists can get it. > But the published `4.2.0-rc.2` on npm you saw in [2] is tagged as "latest" > and have been published to the public. Please remove it, yes that will cause some issues, but the PPMC needs to follow ASF release policy. IMO If you want an exception you will need to talk to the board and/or VP legal to get permission. Thanks, Justin - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Are you talking about the email with subject: "Re: [VOTE] Release Apache ECharts 4.2.1 (release candidate 1)" from 1/29 started by you? I haven't dug in and checked everything like LICENSES and headers or the signatures and but overall it looks really strong and I would vote +1 but you need to have sha256 or sha512 signatures to verify the zip file. I will setup a suitable node.js environment for this so I can test future builds and be more involved. Regards, KAM On 1/30/2019 11:05 AM, Ovilia wrote: > Dear mentors, > > We've published our release vote on general@, but it has got no reply by > now. > Could you have a look at it, and give a +1 perhaps? > > Zhang Wenli > http://zhangwenli.com > > > On Wed, Jan 30, 2019 at 11:34 PM Dave Fisher wrote: > >> Hi - >> >> Also, the project did begin discussions about adding a committer on >> private@. These stalled, but can be restarted now. >> >> Regards, >> Dave >> >> Sent from my iPhone >> >>> On Jan 30, 2019, at 7:24 AM, Kevin A. McGrail >> wrote: >>> Nice response Shuang. If I can help with the voting process for >>> releases, etc., please let me know. >>> >>> I can also try to help with adding more committers, writing notices, >>> etc. to try and get more interest on Social Media even. >>> >>> Regards, >>> KAM On 1/30/2019 3:11 AM, SHUANG SU wrote: Hi Justin, I am afraid it is inevitable to push the rc version to npm. Without >> that, the build process and some local CI process can not be performed, which causes the developers concerning about this project can not test whether the release >> candidate is qualified to be released. But the rc version is not tagged as "latest" in npm, so the public can >> not get this version by `npm install`. So it will not make bad effect, just like publishing >> the rc pack to the Apache svn dev repo [1]. But the published `4.2.0-rc.2` on npm you saw in [2] is tagged as >> "latest" and have been published to the public. That is incorrect according to the Apache >> rule. We should have removed it. But unfortunately we have missed the opportunity to do that. Many outsider projects or systems have depended on that, and the removal will cause >> some disaster. This mistake was made because we did not clear enough about the total release process. But having this experience, this mistake will not happen again. Only the official release will be directed to publish in npm from now on. 1. https://dist.apache.org/repos/dist/dev/incubator/ 2. https://www.npmjs.com/package/echarts Thanks -- Su Shuang (100pah) -- > On Wed, 30 Jan 2019 at 15:14, Justin Mclean >> wrote: > HI, > >> As about "When were the last committers or PPMC members elected?", we > don't have new committers or PPMC members elected after the >> initiation. So > should I fill in NA or the initiation date? > > Yes, but being a year in incubation without adding any committer or PMC > members is concerning, especially given the number of contributors to >> the > project. You might want to put in the report what is being done to >> address > this.There’s some tasks mentioned on the private list that need to be > followed up on by the PPMC I think. > >> As about release version, we tagged "rc" for release candidate >> version. > Does that mean rc version can never be available to the public? > > Release candidates should not be made available to the general public, > they can be made available to people who follow the dev list. [1] They >> can > be placed here. [2] > > Thanks, > Justin > > 1. https://www.apache.org/dev/release-distribution#unreleased > 2. http://www.apache.org/legal/release-policy.html#stage > > > >>> -- >>> Kevin A. McGrail >>> VP Fundraising, Apache Software Foundation >>> Chair Emeritus Apache SpamAssassin Project >>> https://www.linkedin.com/in/kmcgrail - 703.798.0171 >>> >>> >>> - >>> To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org >>> For additional commands, e-mail: dev-h...@echarts.apache.org >>> >> -- Kevin A. McGrail VP Fundraising, Apache Software Foundation Chair Emeritus Apache SpamAssassin Project https://www.linkedin.com/in/kmcgrail - 703.798.0171 - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Dear mentors, We've published our release vote on general@, but it has got no reply by now. Could you have a look at it, and give a +1 perhaps? Zhang Wenli http://zhangwenli.com On Wed, Jan 30, 2019 at 11:34 PM Dave Fisher wrote: > Hi - > > Also, the project did begin discussions about adding a committer on > private@. These stalled, but can be restarted now. > > Regards, > Dave > > Sent from my iPhone > > > On Jan 30, 2019, at 7:24 AM, Kevin A. McGrail > wrote: > > > > Nice response Shuang. If I can help with the voting process for > > releases, etc., please let me know. > > > > I can also try to help with adding more committers, writing notices, > > etc. to try and get more interest on Social Media even. > > > > Regards, > > KAM > >> On 1/30/2019 3:11 AM, SHUANG SU wrote: > >> Hi Justin, > >> > >> I am afraid it is inevitable to push the rc version to npm. Without > that, > >> the build > >> process and some local CI process can not be performed, which causes the > >> developers > >> concerning about this project can not test whether the release > candidate is > >> qualified to be released. > >> But the rc version is not tagged as "latest" in npm, so the public can > not > >> get this version > >> by `npm install`. So it will not make bad effect, just like publishing > the > >> rc pack to the > >> Apache svn dev repo [1]. > >> > >> But the published `4.2.0-rc.2` on npm you saw in [2] is tagged as > "latest" > >> and have been > >> published to the public. That is incorrect according to the Apache > rule. We > >> should have > >> removed it. But unfortunately we have missed the opportunity to do that. > >> Many outsider > >> projects or systems have depended on that, and the removal will cause > some > >> disaster. > >> > >> This mistake was made because we did not clear enough about the total > >> release process. > >> But having this experience, this mistake will not happen again. Only the > >> official release > >> will be directed to publish in npm from now on. > >> > >> 1. https://dist.apache.org/repos/dist/dev/incubator/ > >> 2. https://www.npmjs.com/package/echarts > >> > >> > >> Thanks > >> > >> -- > >> Su Shuang (100pah) > >> -- > >> > >> > >> > >>> On Wed, 30 Jan 2019 at 15:14, Justin Mclean > wrote: > >>> > >>> HI, > >>> > As about "When were the last committers or PPMC members elected?", we > >>> don't have new committers or PPMC members elected after the > initiation. So > >>> should I fill in NA or the initiation date? > >>> > >>> Yes, but being a year in incubation without adding any committer or PMC > >>> members is concerning, especially given the number of contributors to > the > >>> project. You might want to put in the report what is being done to > address > >>> this.There’s some tasks mentioned on the private list that need to be > >>> followed up on by the PPMC I think. > >>> > As about release version, we tagged "rc" for release candidate > version. > >>> Does that mean rc version can never be available to the public? > >>> > >>> Release candidates should not be made available to the general public, > >>> they can be made available to people who follow the dev list. [1] They > can > >>> be placed here. [2] > >>> > >>> Thanks, > >>> Justin > >>> > >>> 1. https://www.apache.org/dev/release-distribution#unreleased > >>> 2. http://www.apache.org/legal/release-policy.html#stage > >>> > >>> > >>> > > > > -- > > Kevin A. McGrail > > VP Fundraising, Apache Software Foundation > > Chair Emeritus Apache SpamAssassin Project > > https://www.linkedin.com/in/kmcgrail - 703.798.0171 > > > > > > - > > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > > For additional commands, e-mail: dev-h...@echarts.apache.org > > > >
Re: ECharts incubator report
Hi - Also, the project did begin discussions about adding a committer on private@. These stalled, but can be restarted now. Regards, Dave Sent from my iPhone > On Jan 30, 2019, at 7:24 AM, Kevin A. McGrail wrote: > > Nice response Shuang. If I can help with the voting process for > releases, etc., please let me know. > > I can also try to help with adding more committers, writing notices, > etc. to try and get more interest on Social Media even. > > Regards, > KAM >> On 1/30/2019 3:11 AM, SHUANG SU wrote: >> Hi Justin, >> >> I am afraid it is inevitable to push the rc version to npm. Without that, >> the build >> process and some local CI process can not be performed, which causes the >> developers >> concerning about this project can not test whether the release candidate is >> qualified to be released. >> But the rc version is not tagged as "latest" in npm, so the public can not >> get this version >> by `npm install`. So it will not make bad effect, just like publishing the >> rc pack to the >> Apache svn dev repo [1]. >> >> But the published `4.2.0-rc.2` on npm you saw in [2] is tagged as "latest" >> and have been >> published to the public. That is incorrect according to the Apache rule. We >> should have >> removed it. But unfortunately we have missed the opportunity to do that. >> Many outsider >> projects or systems have depended on that, and the removal will cause some >> disaster. >> >> This mistake was made because we did not clear enough about the total >> release process. >> But having this experience, this mistake will not happen again. Only the >> official release >> will be directed to publish in npm from now on. >> >> 1. https://dist.apache.org/repos/dist/dev/incubator/ >> 2. https://www.npmjs.com/package/echarts >> >> >> Thanks >> >> -- >> Su Shuang (100pah) >> -- >> >> >> >>> On Wed, 30 Jan 2019 at 15:14, Justin Mclean wrote: >>> >>> HI, >>> As about "When were the last committers or PPMC members elected?", we >>> don't have new committers or PPMC members elected after the initiation. So >>> should I fill in NA or the initiation date? >>> >>> Yes, but being a year in incubation without adding any committer or PMC >>> members is concerning, especially given the number of contributors to the >>> project. You might want to put in the report what is being done to address >>> this.There’s some tasks mentioned on the private list that need to be >>> followed up on by the PPMC I think. >>> As about release version, we tagged "rc" for release candidate version. >>> Does that mean rc version can never be available to the public? >>> >>> Release candidates should not be made available to the general public, >>> they can be made available to people who follow the dev list. [1] They can >>> be placed here. [2] >>> >>> Thanks, >>> Justin >>> >>> 1. https://www.apache.org/dev/release-distribution#unreleased >>> 2. http://www.apache.org/legal/release-policy.html#stage >>> >>> >>> > > -- > Kevin A. McGrail > VP Fundraising, Apache Software Foundation > Chair Emeritus Apache SpamAssassin Project > https://www.linkedin.com/in/kmcgrail - 703.798.0171 > > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Nice response Shuang. If I can help with the voting process for releases, etc., please let me know. I can also try to help with adding more committers, writing notices, etc. to try and get more interest on Social Media even. Regards, KAM On 1/30/2019 3:11 AM, SHUANG SU wrote: > Hi Justin, > > I am afraid it is inevitable to push the rc version to npm. Without that, > the build > process and some local CI process can not be performed, which causes the > developers > concerning about this project can not test whether the release candidate is > qualified to be released. > But the rc version is not tagged as "latest" in npm, so the public can not > get this version > by `npm install`. So it will not make bad effect, just like publishing the > rc pack to the > Apache svn dev repo [1]. > > But the published `4.2.0-rc.2` on npm you saw in [2] is tagged as "latest" > and have been > published to the public. That is incorrect according to the Apache rule. We > should have > removed it. But unfortunately we have missed the opportunity to do that. > Many outsider > projects or systems have depended on that, and the removal will cause some > disaster. > > This mistake was made because we did not clear enough about the total > release process. > But having this experience, this mistake will not happen again. Only the > official release > will be directed to publish in npm from now on. > > 1. https://dist.apache.org/repos/dist/dev/incubator/ > 2. https://www.npmjs.com/package/echarts > > > Thanks > > -- > Su Shuang (100pah) > -- > > > > On Wed, 30 Jan 2019 at 15:14, Justin Mclean wrote: > >> HI, >> >>> As about "When were the last committers or PPMC members elected?", we >> don't have new committers or PPMC members elected after the initiation. So >> should I fill in NA or the initiation date? >> >> Yes, but being a year in incubation without adding any committer or PMC >> members is concerning, especially given the number of contributors to the >> project. You might want to put in the report what is being done to address >> this.There’s some tasks mentioned on the private list that need to be >> followed up on by the PPMC I think. >> >>> As about release version, we tagged "rc" for release candidate version. >> Does that mean rc version can never be available to the public? >> >> Release candidates should not be made available to the general public, >> they can be made available to people who follow the dev list. [1] They can >> be placed here. [2] >> >> Thanks, >> Justin >> >> 1. https://www.apache.org/dev/release-distribution#unreleased >> 2. http://www.apache.org/legal/release-policy.html#stage >> >> >> -- Kevin A. McGrail VP Fundraising, Apache Software Foundation Chair Emeritus Apache SpamAssassin Project https://www.linkedin.com/in/kmcgrail - 703.798.0171 - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
Re: ECharts incubator report
Hi Justin, I am afraid it is inevitable to push the rc version to npm. Without that, the build process and some local CI process can not be performed, which causes the developers concerning about this project can not test whether the release candidate is qualified to be released. But the rc version is not tagged as "latest" in npm, so the public can not get this version by `npm install`. So it will not make bad effect, just like publishing the rc pack to the Apache svn dev repo [1]. But the published `4.2.0-rc.2` on npm you saw in [2] is tagged as "latest" and have been published to the public. That is incorrect according to the Apache rule. We should have removed it. But unfortunately we have missed the opportunity to do that. Many outsider projects or systems have depended on that, and the removal will cause some disaster. This mistake was made because we did not clear enough about the total release process. But having this experience, this mistake will not happen again. Only the official release will be directed to publish in npm from now on. 1. https://dist.apache.org/repos/dist/dev/incubator/ 2. https://www.npmjs.com/package/echarts Thanks -- Su Shuang (100pah) -- On Wed, 30 Jan 2019 at 15:14, Justin Mclean wrote: > HI, > > > As about "When were the last committers or PPMC members elected?", we > don't have new committers or PPMC members elected after the initiation. So > should I fill in NA or the initiation date? > > Yes, but being a year in incubation without adding any committer or PMC > members is concerning, especially given the number of contributors to the > project. You might want to put in the report what is being done to address > this.There’s some tasks mentioned on the private list that need to be > followed up on by the PPMC I think. > > > As about release version, we tagged "rc" for release candidate version. > Does that mean rc version can never be available to the public? > > Release candidates should not be made available to the general public, > they can be made available to people who follow the dev list. [1] They can > be placed here. [2] > > Thanks, > Justin > > 1. https://www.apache.org/dev/release-distribution#unreleased > 2. http://www.apache.org/legal/release-policy.html#stage > > >
Re: ECharts incubator report
HI, > As about "When were the last committers or PPMC members elected?", we don't > have new committers or PPMC members elected after the initiation. So should I > fill in NA or the initiation date? Yes, but being a year in incubation without adding any committer or PMC members is concerning, especially given the number of contributors to the project. You might want to put in the report what is being done to address this.There’s some tasks mentioned on the private list that need to be followed up on by the PPMC I think. > As about release version, we tagged "rc" for release candidate version. Does > that mean rc version can never be available to the public? Release candidates should not be made available to the general public, they can be made available to people who follow the dev list. [1] They can be placed here. [2] Thanks, Justin 1. https://www.apache.org/dev/release-distribution#unreleased 2. http://www.apache.org/legal/release-policy.html#stage
Re: ECharts incubator report
Hi Justin, I updated the report and here are some questions about it. As about "When were the last committers or PPMC members elected?", we don't have new committers or PPMC members elected after the initiation. So should I fill in NA or the initiation date? As about release version, we tagged "rc" for release candidate version. Does that mean rc version can never be available to the public? That would make the whole process too long for a new release. Is there anything else we could do? For the other problems, hopefully, I've updated in the report. Zhang Wenli http://zhangwenli.com On Wed, Jan 30, 2019 at 4:13 AM Justin Mclean wrote: > Hi, > > Thanks for submitting a report however it's a little light on detail and > missing some information. Could you please update this. > > Can you think of something better than "3. Fixing bugs and improving > features" as a step towards graduation? What does the project need to aim > for in order to graduate? > > I see you have put N/A under "When were the last committers or PPMC > members elected?" I don't believe this is correct. > > You are also missing information on the current release vote. > > While there's no requirement to do so, I find that you get a better > quality report if it is discussed on the list and is a collaborative > effort. Perhaps try this next time? > > It's also come to my attention that you have unproved releases here [1], > the PPMC can't release any software to the general public that hasn't been > approved for release. Please remove those releases. Please mention what you > are going to doing to address this in your report. > > Thanks, > Justin > > P.S I am not subscribed to this list so please CC me on any replies. > > 1. https://www.npmjs.com/package/echarts > > - > To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org > For additional commands, e-mail: dev-h...@echarts.apache.org > >
ECharts incubator report
Hi, Thanks for submitting a report however it's a little light on detail and missing some information. Could you please update this. Can you think of something better than "3. Fixing bugs and improving features" as a step towards graduation? What does the project need to aim for in order to graduate? I see you have put N/A under "When were the last committers or PPMC members elected?" I don't believe this is correct. You are also missing information on the current release vote. While there's no requirement to do so, I find that you get a better quality report if it is discussed on the list and is a collaborative effort. Perhaps try this next time? It's also come to my attention that you have unproved releases here [1], the PPMC can't release any software to the general public that hasn't been approved for release. Please remove those releases. Please mention what you are going to doing to address this in your report. Thanks, Justin P.S I am not subscribed to this list so please CC me on any replies. 1. https://www.npmjs.com/package/echarts - To unsubscribe, e-mail: dev-unsubscr...@echarts.apache.org For additional commands, e-mail: dev-h...@echarts.apache.org
