[Incubator] Recommend Graduation of Rasdaman from OSGeo Incubation
Cameron Shorter
cameron.shorter at gmail.com
Sun Jan 25 03:11:27 PST 2015
On 23/01/2015 4:33 am, Peter Baumann wrote:
> Cameron-
>
> coming back on this, after some delay unfortunately.
> Responses inline:
>
> On 10/31/2014 08:57 PM, Cameron Shorter wrote:
>> Thanks for the responses Peter, and offering to put the final touches
>> on your incubation report. Comments in line.
>>
>> On 1/11/2014 2:07 am, Peter Baumann wrote:
>>> Hi Cameron,
>>>
>>> see inline for the individual responses:
>>>
>>> On 10/18/2014 11:30 PM, Cameron Shorter wrote:
>>>> Hi Peter and Bruce,
>>>> I've reviewed the incubation documentation provided, and provided a
>>>> few spot reviews of the referenced documentation.
>>>> Overall, I think the documentation is in good shape, but missing
>>>> explanations, particularly around licenses and testing.
>>>>
>>>>
>>>> Below are a few things that I'd like to see looked into before
>>>> final sign off:
>>>>
>>>> Looking at: http://rasdaman.org/wiki/OSGeoIncubationChecklist which
>>>> provides the summary about how the project is ready to complete
>>>> graduation.
>>>>
>>>> * For each statement that you are addressing, please add a brief
>>>> sentence explaining HOW you are addressing the statement, possibly
>>>> also with an OPINION.
>>>> For instance, currently for: "Open communication channels", you
>>>> have "see MailingLists".
>>>> In reviewing, I can see that a mailing list has been set up. In
>>>> looking at the list, I see users are asked questions on the user
>>>> list and there are answers. I can see that auto-patch messages are
>>>> being pumped into the developers list. In my quick scan, I can't
>>>> see discussions between developers about a new feature, and
>>>> discussions about a design idea. So I'd suggest describing how
>>>> communication is happening. Maybe:
>>>> "See MailingLists <URL>. User questions are typically addressed on
>>>> the user email list. Development is currently mostly happening at
>>>> Jacobs University, and discussions about development typically
>>>> happen in person, and the results of such conversations are
>>>> summarised onto the developer email list (or ...)."
>>>
>>> OK, we will try adding something I have never seen such a thing on
>>> any of the graduation reports, and neither is this required by the
>>> questionnaire.
>>> Hence, can you help us by providing an example project showing what
>>> OSGeo expects?
>>
>> I've just had a look at
>> http://wiki.osgeo.org/wiki/GeoTools_Project_Status
>> In some places the added description is light. Eg description about
>> how email lists is used is brief. However, if I follow the link to
>> the list it is obvious how well the lists are being used.
>>
>> A good example for you to follow re comments from the GeoTools doc is:
>>
>> How many active developers are there? Are they from multiple
>> organizations?
>> There are around 10 developers working on geotools as part of
>> their day job, the expanded list goes up to around 30 with commit
>> access. The important part is the number of responsible module
>> maintainers. While the project is made up of individuals (who
>> happen to work for organizations such as Penn State University,
>> Refractions Research, The Open Planning Project, IBM, Oracle,
>> Google others...)
>
> see answer to next post, following your recommendation there.
Peter,
You have not addressed my comment.
To be clear, EVERY incubation checkpoint needs to be addressed, by
providing a DESCRIPTIVE answer explaining HOW the requirement is
addressed, along with applicable REFERENCE(S) to supporting evidence.
At the moment, almost all points on your checklist only provide a
REFERENCE, and don't provide a DESCRIPTION.
To avoid misunderstanding, you should be updating almost ALL points, not
just the points I've expanded upon below.
Until the Rasdaman community addresses this, I'd consider the incubation
process incomplete.
>
>
>>
>>
>>>
>>>>
>>>> * For "Open Source License": I'd be hoping to see a statement about
>>>> the dual license, including a comment about the enterprise license.
>>>> I'm ok with the dual license concept which I think fits with OSGeo
>>>> principles. Further, from the material and links provided, it
>>>> seems plausible that the GPL license terms are broken by LGPL and
>>>> enterprise licensed products being distributed.
>>>> I feel there needs to be an explanation of the enterprise license,
>>>> and then about how LGPL products and enterprise products don't need
>>>> to be GPL (because they don't link in any GPL libraries?).
>>>
>>> Available - there is even a dedicated page on this, just follow the
>>> link:
>>> http://rasdaman.org/wiki/License
>>
>> The above link references enterprise features, but doesn't reference
>> an enterprise license. If you are referencing "enterprise rasdaman" I
>> would expect to be able to find a read the enterprise license.
>
> here it is, on said page:
>
> * A commercially supported, non-GPL-restricted variant of rasdaman
> <http://www.rasdaman.com> is available which we recommend to all
> users who have requirements that go beyond GPL.
>
Peter,
The above link points to http://www.rasdaman.com/ which doesn't contain
license text.
Did you not understand the question? I was suggesting that the license
page should reference the commercial license, which might look something
like: http://www.adobe.com/legal/general-terms.html
If you don't have such license text available, you could say "for more
details about the commercial version, contact aaa at bbb.com".
Note: This is a suggestion which I think should be in place for a
professional product, but as it doesn't apply to the open source version
of Rasdaman, I would accept an argument that this statement is not
applicable to OSGeo incubation.
>
>>>
>>>>
>>>> * "/Long term viability/":
>>>> Sentence needed here discussing roughly how many contributors you
>>>> have, (as well as existing reference to patchmanager).
>>>
>>> Patchmanager is the source for the information on this, just choose
>>> "all" to see all contributors.
>>> Hm, not sure what else would need to be added.
>>
>> I suggest adding a statement like this:
>>
>> /There are around 10 developers working on geotools as part of their
>> day job, the expanded list goes up to around 30 with commit access.
>> The important part is the number of responsible module maintainers.
>> While the project is made up of individuals (who happen to work for
>> organizations such as Penn State University, Refractions Research,
>> The Open Planning Project, IBM, Oracle, Google others...)/
>>
>> and add
>> /for further details, refer to matchmanager, and select all./
>
> Part 1 is varying and cannot be maintained upto date;
> as for part 2, this exists already:
> see contributors in http://rasdaman.org/patchmanager
Peter, it seems you haven't made any changes to address my comment?
I can't see text like that I suggested above.
You are not describing how you meet a graduation criteria. You need to
help a reviewer understand how to interpret patchmaker by summarising
the results in a statement.
>
>
>>>
>>>>
>>>> * /Project documentation is available under an open license, as
>>>> documented here: Legal/:
>>>> The referenced "Legal" link doesn't talk about Documentation.
>>>> Please reference correct documentation license link.
>>>
>>> oops, that should read "License" - fixed now. Good catch!
Thanks, looks good now.
>>>
>>>>
>>>> * /http://rasdaman.org/wiki/Legal/ :
>>>> Remove references to "... blah-blah ..."
>>>
>>> why? ;-) cannot see much relevance for incubation as long as even
>>> industry does not complain. But we are getting lost in details,
>>> admittedly.
>> I assume you are making a joke, but in case you are not, I'll explain.
>>
>> The line which references "blah-blah" looks like it comes from a
>> template and hence should be removed in a final document. And in
>> particular, professional organisations (which is one of the user
>> groups that OSGeo targets), expects to have professional level legal
>> documentation which doesn't reference "blah-blah".
>
> sure, corrected now :)
thanks.
>
>
>>
>>>
>>>>
>>>> * " /Which includes execution of the testing process before
>>>> releasing a stable release: http://212.201.49.21:35000/"/ :
>>>> Link requires user log in. A sentence explaining testing will help
>>>> here.
Still not actioned. I'm looking for a sentence explaining how the
testing process works.
>>>
>>>>
>>>> * "/All patches submitted to the repository undergo review before
>>>> being applied to the code base:
>>>> http://www.rasdaman.org/patchmanager/" :
>>>> This doesn't confirm that patches are reviewed (by a person), only
>>>> that you use patchmanager.
>>>
>>> ok, added "human".
>>
>> Checked http://rasdaman.org/wiki/OSGeoIncubationChecklist, I can't
>> see a reference to a process which includes reviewing patches before
>> they are applied.
>
> see Governance:
> Patches are accepted by members of the PSC who all have commit rights.
Ok, good. Only problem is that person reading the checklist:
http://rasdaman.org/wiki/OSGeoIncubationChecklist can't find it. For
each question, including this, you need a sentence stating how you
address the question, quoting text to your procedures, and providing a
cross reference, in this case I'd quote the words above, and provide a
link to the Governance document.
>
>>>
>>>>
>>>> * "/Systemtests are to be run by developers before submitting
>>>> patches: http://www.rasdaman.org/wiki/CodeGuide/" :
>>>> This code guide is good, but on my first pass, I can't see any
>>>> reference to testing processes. How do you verify that all code is
>>>> tested? Ideally reference continual integration or similar.
>>>
>>> We verify automatically by running a test against each patch; the
>>> outcome is documented next to each patch once the test run has
>>> finished. This is quite a new feature, though, and - as you have
>>> spotted - not all is in place yet.
>>> The page is: http://www.rasdaman.org/wiki/RasdamanTestSuites
>>>
>> OK, this test wiki you reference is a good start, but I notice it is
>> still in draft state, with "TBD" and "TODO" text in it.
>>
>> This would need to be complete before I'd consider rasdaman is ready
>> to complete incubation.
>
> Hm, are really all graduated projects without any TBDs etc? I found
> TBDs on osgeo.org, and 455 TODOs on PostGIS. How would you think about
> that in comparison?
>
> Work anyway will progress in due course and depending on project
> situation, it's in our own vest interest.
Some TBDs are more important than others.
OSGeo is a stamp of quality, and a core component of quality is to have
established testing.
As an incubation committee member, I'd vote -1 on incubating a project
that has TBDs in their quality processes.
I suggest addressing this.
>
>
>>
>> I suggest that this test wiki should also be referenced from:
>> http://www.rasdaman.org/wiki/CodeGuide
>
> Cameron, somehow I feel we are getting into micro management. If we
> look at incubated projects we find a range of different situations,
> and we do not even speak about regular checking whether incubation
> quality is maintained.
> While I am open for hints on improvement I am not sure that such
> finetuning has to do with the original goals of incubation.
>
> So let's see where we stand in year 5 since rasdaman's incubation
> request :)
Peter,
The review I've provided for you is comparable to reviews I've provided
to most of the other OSGeo incubated projects. For all these projects,
the project owners quickly addressed the comments, and soon after, they
were incubated.
I suggest you don't imply blame for Rasderman's lack of incubation. You
might not like who gets implicated.
Yes, incubation is hard. People recognise the effort and level of
quality required to pass OSGeo-Incubation. It is why these projects are
favoured when organisations are selecting software.
Rasdaman is close to addressing all incubation criteria. Lets see if we
can get it across the line.
>
> cheers,
> Peter
>
>
>>
>>>>
>>>> * Link to /http://rasdaman.org/browser/systemtest/ is broken. I'm
>>>> looking for evidence of testing processes, including unit testing,
>>>> continuous integration, and a system testing plan which is executed.
>>>
>>> sorry for the broken link, fixed that:
>>> http://www.rasdaman.org/wiki/RasdamanTestSuites
>> thanks
>>
>>>
>>> HTH,
>>> Peter
>>>
>>>>
>>>>
>>>>
>>>> On 16/10/2014 4:25 pm, Bruce Bannerman wrote:
>>>>>
>>>>> Project: Rasdaman
>>>>>
>>>>> Project Contact: Peter Baumann
>>>>>
>>>>> Project Mentor: Bruce Bannerman
>>>>>
>>>>>
>>>>> Recommendation: Rasdaman to Graduate from OSGeo Incubation
>>>>>
>>>>>
>>>>> Incubation Checklist:
>>>>> http://rasdaman.org/wiki/OSGeoIncubationChecklist
>>>>>
>>>>>
>>>>>
>>>>> Comments:
>>>>>
>>>>> I have been monitoring and participating in the Rasdaman Community
>>>>> developer and user email lists for over four years. I plan to
>>>>> continue to participate within the community.
>>>>>
>>>>> It has been pleasing to see how the community has evolved over
>>>>> this time to develop effective, open and robust processes.
>>>>>
>>>>> I recommend to other FOSS4G communities to review what has been
>>>>> done, as you may well find that Rasdaman offers a compelling
>>>>> solution for the management of multi-dimensional array or gridded
>>>>> data that your project may be able to capitalise on.
>>>>>
>>>>> A summary of community activity may be found at:
>>>>> https://www.openhub.net/p/rasdaman
>>>>>
>>>>> Key project members are active within both OSGeo and the OGC.
>>>>>
>>>>> Within the OGC they have successfully guided a range of standards
>>>>> to support mutltidimensional array (gridded) data which have
>>>>> culminated (currently) in Web Coverage Service version 2.0 and a
>>>>> suite of extensions, see: http://www.opengeospatial.org/standards/wcs
>>>>>
>>>>> The project's principal, Professor Peter Baumann was also recently
>>>>> awarded OGC's prestigious Kenneth D Gardels award for his
>>>>> contributions to OGC.
>>>>>
>>>>>
>>>>>
>>>>> After having participated within the community for four years and
>>>>> having reviewed their OSGeo Project Graduation Checklist, I
>>>>> believe that Rasdaman Community have fulfilled the requirements of
>>>>> OSGeo Incubation and are eligible to graduate as an OSGeo Project.
>>>>>
>>>>>
>>>>> The only reservation that I have is over the dual licensing model
>>>>> that is used by the sponsoring company, Rasdaman GmbH. See:
>>>>> http://www.rasdaman.com/Product/commercial+free.php
>>>>>
>>>>>
>>>>> There are two versions: Community (GPL and LGPL) and Enterprise
>>>>> (free from GPL constraints).
>>>>>
>>>>> The main difference between the two versions is summarised at:
>>>>> http://www.rasdaman.org/wiki/Features
>>>>>
>>>>>
>>>>> This dual licensing may create a problem for the Rasdaman
>>>>> Community in the future should Rasdaman GmbH decide to sell its
>>>>> stake in the Rasdaman Enterprise version of the software. We have
>>>>> already seen a similar situation arise after the sale of MySQL
>>>>> which was similarly licensed.
>>>>>
>>>>> However as the community edition of the software is released under
>>>>> open source licenses, I expect that the main impact would be to
>>>>> the continuity of the Rasdaman Community in its current form. In
>>>>> the event of a sale, we may see the software forked, as happened
>>>>> with MySQL.
>>>>>
>>>>> For consideration and vote.
>>>>>
>>>>> Bruce
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Incubator mailing list
>>>>> Incubator at lists.osgeo.org
>>>>> http://lists.osgeo.org/mailman/listinfo/incubator
>>>>
>>>> --
>>>> Cameron Shorter,
>>>> Software and Data Solutions Manager
>>>> LISAsoft
>>>> Suite 112, Jones Bay Wharf,
>>>> 26 - 32 Pirrama Rd, Pyrmont NSW 2009
>>>>
>>>> P +61 2 9009 5000, Wwww.lisasoft.com, F +61 2 9009 5099
>>>
>>> --
>>> Dr. Peter Baumann
>>> - Professor of Computer Science, Jacobs University Bremen
>>> www.faculty.jacobs-university.de/pbaumann
>>> mail:p.baumann at jacobs-university.de
>>> tel: +49-421-200-3178, fax: +49-421-200-493178
>>> - Executive Director, rasdaman GmbH Bremen (HRB 26793)
>>> www.rasdaman.com, mail:baumann at rasdaman.com
>>> tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
>>> "Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)
>>>
>>>
>>
>> --
>> Cameron Shorter,
>> Software and Data Solutions Manager
>> LISAsoft
>> Suite 112, Jones Bay Wharf,
>> 26 - 32 Pirrama Rd, Pyrmont NSW 2009
>>
>> P +61 2 9009 5000, Wwww.lisasoft.com, F +61 2 9009 5099
>
> --
> Dr. Peter Baumann
> - Professor of Computer Science, Jacobs University Bremen
> www.faculty.jacobs-university.de/pbaumann
> mail:p.baumann at jacobs-university.de
> tel: +49-421-200-3178, fax: +49-421-200-493178
> - Executive Director, rasdaman GmbH Bremen (HRB 26793)
> www.rasdaman.com, mail:baumann at rasdaman.com
> tel: 0800-rasdaman, fax: 0800-rasdafax, mobile: +49-173-5837882
> "Si forte in alienas manus oberraverit hec peregrina epistola incertis ventis dimissa, sed Deo commendata, precamur ut ei reddatur cui soli destinata, nec preripiat quisquam non sibi parata." (mail disclaimer, AD 1083)
>
>
--
Cameron Shorter,
Software and Data Solutions Manager
LISAsoft
Suite 112, Jones Bay Wharf,
26 - 32 Pirrama Rd, Pyrmont NSW 2009
P +61 2 9009 5000, W www.lisasoft.com, F +61 2 9009 5099
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/incubator/attachments/20150125/3c26b480/attachment-0001.html>
More information about the Incubator
mailing list