[Incubator] Oskari Incubation status

Thu Aug 23 07:46:58 PDT 2018

Dear all,

coming back to this topic after we’ve had a chance to discuss this within the Oskari community.

There are at least two things where we see that we should (and hopefully can) improve:
1) Openness of our communication
2) Reducing the risk factor of the biggest contributor abandoning the project (or broader adoption)

The communication thing is a bit tricky, we have the email list and its usage is growing, which is obviously good. We are trying to steer people into using the mailing list for the reasons already mentioned (archiving and ability to search for old messages). Then we have our Slack instance for quicker communication and that has at least to issues: At this point it still needs an invitation that has to be sent manually, there is a solution for this and probably in the near future we can automate this task. Then the free tier Slack only archives 10k messages so we need a way to archive them ourselves. I know Slack supports exports, but we have to look into it in a bit more detail. Sending the archive to the email list once a week or once a day might be good and shouldn’t be that difficult.

Another thing is that most of the discussion happens in Finnish as two Finns are communicating. Do you think this is a problem? Of course we can ask people to communicate in English, but I’m afraid that will lead to more direct/private messages being used and they will not show in the archives. I know many other FOSS4G-projects have a lot of communication in other languages than English, so probably I shouldn’t worry?

Then about the risk of NLSFI abandoning the project, or getting broader adoption for our software. There are probably things that we have failed to communicate properly, we have a feeling our community is developing at a nice pace and actually doing more things together that is actually visible “from the outside”. So in addition to growing the community we need to improve our communication as well. Last year the board of the joint development forum, which consists of nine organizations jointly made a request for tender for a “community coordinator” for the Oskari community. This was done to improve our communication and co-operation and also to reduce the role of NLSFI in the community-owned project. Are there some concrete steps/hints that come to your mind? What kind of things would you recommend? We’re all ears :)

One more thing: Arnulf mentioned in the earlier response that “I cannot find a good indicator supporting that there is an open community and communication.”. What kind of concrete indicators would you like to see? What could be a good goal?

And again, thank you for your answers and help. Getting it from people with your level of experience means a lot to us.

Cheers,
Timo

PS. Sami and me will be traveling to Dar es Salaam, come and have a chat with us if you’re also coming!

From: Arnulf Christl (Metaspatial) [mailto:arnulf.christl at metaspatial.net]
Sent: 27. heinäkuuta 2018 12:57
To: Cameron Shorter <cameron.shorter at gmail.com>; incubator at lists.osgeo.org
Cc: Aarnio Timo (MML) <timo.aarnio at maanmittauslaitos.fi>; Sanna Jokela (Gispo) <sanna at gispo.fi>
Subject: Re: [Incubator] Oskari Incubation status

Cameron,

thanks for the pointer to the new / old/ checklist link.

All,

the Incubation process and Checklists are not in a good state right now. There also have been changes during the migration from the old web site to the new website. For example the topic of requiring Open Source / Free Software licenses is much less prominent than on the old linked version Cameron provided and the even older Incubation Status pages Oskari had started on. I am pretty sure that these changes have not been discussed on the Incubation lists?! The official document on the web (https://www.osgeo.org/resources/project-graduation-checklist/) points to a PDF file. Why on earth would we want this to be a file and not plain HTML?! Incubation is core and heart to OSGeo and the content is not readily available on the web?

No dramatic thing but we may want to go through the whole process on this list again, flesh out old docs and so on. Happy to help out on this around October / November when my schedule relaxes a bit. If anybody is eager to push this forward before then please go ahead.

Over the coming days we will move the Oskari Incubation status page content to the version provided by Cameron and then ask the Incubation committee to comment and eventually come to a decision whether we can recommend Oskari for graduation.

It is somewhat hard to find things on the OSGeo Web in general because many links from the Wiki to the new website are broken. This causes quite a disconnect. When trying to fix some of the links I also noticed difference in content and language which we may have to fix with a lot of manual editing so that Wiki and website are in sync again. Would be great to see this coordinated by the web site committee - but that channel is dead since December 2017 which is unfortunate because we now have a nice web site but little editing going on - but this is a another topic.

Have fun,
Arnulf

Am 2018-07-08 um 10:01 schrieb Cameron Shorter:

Aarnio, Arnulf,

I just checked up on the link Arnulf pointed us to, https://wiki.osgeo.org/wiki/Oskari_Incubation_Status, and it appears you are using an old version of the OSGeo Incubation Checklist. This is understandable, as our incubation docs have not been copied across into our new website. (Mumble, mumble, grrr, grr.)

This is the checklist we should be using:
http://old.www.osgeo.org/incubator/process/project_graduation_checklist.html

On 6/7/18 10:10 pm, Aarnio Timo (MML) wrote:
Dear list members!

Thank you so much for you assesment Arnulf and thank you for your valuable comments Cameron and Bruce!

I’ll try to answer some of the questions and concerns raised here.

The mailing list has been so far more as Arnulf described, an announcement channel. However there have been multiple questions and answers regarding the software as well. You’ve gotta start somewhere right? Also some people seem to dislike email lists and say that they are old fashioned. Personally I don’t completely agree but I see where these comments are coming from.

In our day to day work we use Slack (and nowadays also Rocket.Chat) for communication. I agree that having logs from our chats might be beneficial for others, but there are at least two things to note here:
- We use the free version of Slack which supports only 10k messages of history, so we’d have to develop some method of fetching the chat archive (+attached media) and then store&publish it somewhere. Not sure what the terms of use for Slack say about this - we could check.
- A lot of the discussion (chat) is in Finnish. Also a lot of it happens via private messages between two developers and even though we’ve tried to encourage people to chat on the public channels sometimes it just does not happen.

Anyone that has questions and/or is developing something to Oskari is welcomed to our Slack. We even planned to automate the invitations, so that one wouldn’t have to ask one from us.

For now I guess the best source for discussions and decisions is GitHub and the PSC meeting notes. We also planned to make the PSC meetings happen regularly (e.g. once a month) in the sake of sharing information & knowledge exchange.

We’ve discussed the OSGeo Live issue early last year and decided that at this point it’s not something we will focus on. Like Arnulf noted Oskari is a “platform” product and relies heavily on external data sets so it’s not that well suited for OSGeo Live.

About the question about Roadmap items and external contributors: Oskari has had external contributors for quite some time, but obviously here more is better. The most recent “external” contribution is this https://github.com/oskariorg/oskari-docs/issues/61 - as an example of something that “went through the process”. I put the word external in quotes because I don’t know what qualifies as being external? The free-to-join Oskari Community has now more than 30 organizations and many of them are using or developing Oskari actively. Last year the first company (then Dimenteq Ltd now Sitowise Ltd) started selling Oskari as a service and also Oskari development and maintenance. This year another company (Gispo Ltd) started offering crowd-funding facilitating services for developing Oskari collaboratively - tbh they are piloting it for now. I really see both of these examples as very positive development and I think they also secure the future of Oskari to some extent.

Obviously a more broader adoption is something we still strive for. For a software like Oskari it’s challenging but little by little we’ve already managed to do it so it’s possible :)

I don’t think we’re in any rush for the graduation and this discussion will be very helpful for us to develop our process even further. So please do comment back so we know what kind of steps to take next and how to proceed. My holidays start today so don’t wonder if I’m not replying for a while, I’ll get back to work in August.

Thanks again for your support!

BR,
Timo

From: Incubator [mailto:incubator-bounces at lists.osgeo.org] On Behalf Of Bruce Bannerman
Sent: 6. heinäkuuta 2018 0:17
To: Cameron Shorter <cameron.shorter at gmail.com><mailto:cameron.shorter at gmail.com>
Cc: Oskari-user at lists.osgeo.org<mailto:Oskari-user at lists.osgeo.org>; Sanna Jokela (Gispo) <sanna at gispo.fi><mailto:sanna at gispo.fi>; incubator at lists.osgeo.org<mailto:incubator at lists.osgeo.org>
Subject: Re: [Incubator] Oskari Incubation status

Hi Arnulf,

Thanks for your detailed assessment. ...and yes, I did read to the bottom ;-)

I would like to see a response to your questions from Timo, or another representative of the Oskari community. In particular a response to the issues of open communities and decision making.

From what I’ve viewed on the project website, Oskari has considerable potential. However, good software is only part of the picture.

Having a good, strong, healthy and active community that is diverse and uses open processes will help the project to survive when funding from the major sponsor ceases.

Kind regards,

Bruce

On 6 Jul 2018, at 03:02, Cameron Shorter <cameron.shorter at gmail.com<mailto:cameron.shorter at gmail.com>> wrote:

Arnulf, to answer your question at the bottom, yes I read to the bottom, and really appreciate your detailed assessment.

Based on Arnulf's assessment, it sounds like Sskari probably has a good codebase, with good work from within one organisation, but is yet to attract engaged external developers. This is a great milestone, and worth acknowledging. But the "unicorn" projects that we look for in OSGeo incubation are those with a healthy community from multiple organisations.

The "extent of collaboration" criteria provides validation of the quality of the project, and is a very good indicator of long term sustainability.
To the Oskari folks, you might want to read this presentation I gave a while back on business justifications for backing collaboration:
http://cameronshorter.blogspot.com/2017/11/open-source-pitch-for-your-boss.html

Without digging into the details any further, I suspect Oskari would need to work on attracting co-contributors, from multiple organisations, and we would likely see evidence of this happening by seeing archive logs of collaborative email lists, or slack or IRC or similar.

Warm regards, Cameron
On 6/7/18 2:11 am, Arnulf Christl (Metaspatial) wrote:

Hi Folks,

this is an update on the Incubation process of Oskari.

https://wiki.osgeo.org/wiki/Oskari_Incubation_Status

As mentor I am still somewhat hesitant to recommend Oskari for OSGeo Incubation because I cannot find a good indicator supporting that there is an open community and communication. Personally I know some of the current core team and totally trust them to work in the Open Source way of doing things. But this may not be quite apparent to anybody not into the project.

Oskari is doing a good job posting news and updates to the OSGeo user list (https://lists.osgeo.org/pipermail/oskari-user/) but the list is not really used by users for any kind of discussion, help or future development. It feels more like a low volume announce list (which is totally OK, so please keep it up).

How and where is the development being discussed? Who can join the Oskari Slack channel - or whatever else is used to have technical and strategic discussions? Are there archives of discussions and decisions? Do you manage to do this in English or is it still everything communicated in Finnish?

In a PSC presentation from December 2016 one bullet-point says that Oskari may become a project on OSGeo live: https://oskariorg.github.io/files/20161220-Oskari_OSGEO_PSC.pdf.
<https://oskariorg.github.io/files/20161220-Oskari_OSGEO_PSC.pdf>

Actually, Incubation is not a requirement, so please feel free to approach the OSGeo Live team to find out more.

In the Roadmap document (which in my opinion is actually ery good) at: https://github.com/oskariorg/oskari-docs/wiki/Roadmap-process you say that:

"After communicating with Oskari community the roadmap items enter Active stage [...]".

Can you give us an example of where and how this has happened?

There are some updates on GitHub issues but it appears to be mostly internal team members (which is totally OK, is this where the collaborative aspect of the future development takes place?

https://github.com/oskariorg/oskari-docs/labels/roadmap

Regarding Code Copyright Review you state that:

All code has been developed by the registered developers listed on github who have signed the CLA. All external libraries have project compatible licenses. The project has been started as a regular Open Source project following the guidelines as set out by OSGeo. A file-per-file code review was therefore deemed superfluous.

As mentor I can confirm that this is correct and satisfies OSGeo's requirements (and yes, I did some quick checks on random code files but due to the well organized origin of the project did not go through all code like we had to in the GRASS project... :-).

In the last PSC meeting in December 2017 a decision was taken to schedule the next meeting "as needed".

Meetings/activity

  *   [...]
  *   future meetings schedule: continue with “as needed” (no need for a fixed schedule)

Summarizing the current status: We discussed some of the above topics before and it is not a requirement for OSGeo Incubation to have active mailing lists and so on. But it is a requirement to have an open process and ideally also a somewhat growing user community. Obviously due its character as a complex portal platform Oskari is not a software that will have millions of downloads but a little broader adoption would be nice to see.

You also say:

  *   Slack/Mailinglist can be used more actively to discuss any issues/voting

If you are not using the Slack channel (which unfortunately is not open), where is the communication taking place?

The PSC meeting notes from December 2017 has a list of current installations / users / contributions. This is good to see. Has there been any additions, changes over the past half year?

Even if you can not produce evidence for any or all of the above issues I will be happy to recommend Oskari for Incubation if you believe that this will help the project. Then it will be up to the Incubation committee to decide whether we can recommend graduation to the OSGeo Board of Directors.

We will have to expect some discussion on the Incubation list on the above topics and this is not bad but a sign of a healthy process.

Internal note to the Incubation committee

IF you ever made it down to this line and maybe even checked some of the referenced documents, please be so kind as to acknowledge this on the mailing list so that we can help Oskari move on.

It may also be a good idea to go back to the Incubation process and check whether it should be revised for certain aspects (especially requirement on mailing lists / alternatives like Slack).

Anything else you need, let me know.

Thank you,

Arnulf

Am 2018-06-25 um 17:05 schrieb Aarnio Timo (MML):
Hi Incubator-list!

We’d like to inquire about the incubation status of Oskari. What are the next steps should be in the incubation process? In our end we’ve done everything we’ve realized that we have to do. But is there something still wanted from us? Or is the process in the OSGeo end now? If so, what are the steps there and is there possibly any estimate when we can think about graduating?

Kind regards,
Timo Aarnio

_______________________________________________

Incubator mailing list

Incubator at lists.osgeo.org<mailto:Incubator at lists.osgeo.org>

https://lists.osgeo.org/mailman/listinfo/incubator

--

Spatially enabling your business

http://metaspatial.net

_______________________________________________

Incubator mailing list

Incubator at lists.osgeo.org<mailto:Incubator at lists.osgeo.org>

https://lists.osgeo.org/mailman/listinfo/incubator

--

Cameron Shorter

Technology Demystifier

Open Technologies and Geospatial Consultant

M +61 (0) 419 142 254
_______________________________________________
Incubator mailing list
Incubator at lists.osgeo.org<mailto:Incubator at lists.osgeo.org>
https://lists.osgeo.org/mailman/listinfo/incubator

--

Cameron Shorter

Technology Demystifier

Open Technologies and Geospatial Consultant

M +61 (0) 419 142 254

_______________________________________________

Incubator mailing list

Incubator at lists.osgeo.org<mailto:Incubator at lists.osgeo.org>

https://lists.osgeo.org/mailman/listinfo/incubator

--

Spatially enabling your business

http://metaspatial.net
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/incubator/attachments/20180823/85c4ae1d/attachment-0001.html>