Re: Help on Data Versioning section

Hoi,

Thanks for the great work Annette ! +1 to suggest SemVer in the
implementation of the best practice z for APIs.
The BPs you suggest are well explained and relevant but I think it would be
best not to get into the schema VS instances discussion.
We can argue that every resource behind a URI can be versioned and
recommend ways to do so.

Furthermore we may want to relate these BPs with the ones about
preservation as a indication on how to keep past versions.

Christophe

On 15 January 2015 at 02:12, Annette Greiner <amgreiner@lbl.gov> wrote:

> copying the dwbp list on this thread, in case anyone would like to see
> where we are with the versioning BPs.
> -AG
> --
> Annette Greiner
> NERSC Data and Analytics Services
> Lawrence Berkeley National Laboratory
> 510-495-2935
>
> On Jan 14, 2015, at 5:10 PM, Newton Calegari <newton@nic.br> wrote:
>
> Cool!
>
> I think our thread is going well.
> It would be nice share it with the whole group?
> If so, how can we do that, just to copy dwbp-mail on this thread or to
> write a new mail describing the main topics discussed?
>
> Newton
>
> Em 14/01/2015, à(s) 23:05, Annette Greiner <amgreiner@lbl.gov> escreveu:
>
> +1 from me!
> --
> Annette Greiner
> NERSC Data and Analytics Services
> Lawrence Berkeley National Laboratory
> 510-495-2935
>
> On Jan 14, 2015, at 4:28 PM, Newton Calegari <newton@nic.br> wrote:
>
> Ops, I haven’t sent the link referencing SchemaVer [1].
>
> About the number versioning, I liked the way proposed in SchemaVer [2]:
>
> Given a version number MODEL-REVISION-ADDITION, increment the:
>
>    - MODEL when you make a breaking schema change which will prevent
>    interaction with *any* historical data
>    - REVISION when you make a schema change which *may* prevent
>    interaction with *some* historical data
>    - ADDITION when you make a schema change that is compatible with *all*
>    historical data
>
> Syntactically this feels similar to SemVer - but as you can see from the
> increment rules, the semantics of each element are very different from
> SemVer.
>
> Is it ok if we propose to use it?
>
> cheers,
> Newton
>
> [1]:
> http://snowplowanalytics.com/blog/2014/05/13/introducing-schemaver-for-semantic-versioning-of-schemas/
> [2]:
> http://snowplowanalytics.com/blog/2014/05/13/introducing-schemaver-for-semantic-versioning-of-schemas/#schemaver
>
>
>
> Em 14/01/2015, à(s) 19:10, Annette Greiner <amgreiner@lbl.gov> escreveu:
>
> Oh, that is awesome. I was initially thinking about trying to use SemVer,
> but of course it didn’t really makes sense.
> p.s., I am generally available via gchat (annettegreiner@gmail.com)
> or Skype (annette.greiner.1) with some advance warning.
> --
> Annette Greiner
> NERSC Data and Analytics Services
> Lawrence Berkeley National Laboratory
> 510-495-2935
>
> On Jan 14, 2015, at 12:56 PM, Newton Calegari <newton@nic.br> wrote:
>
> Hi Annette,
>
> Thanks for your contribution! It is a big and nice work!
>
> I’ve one just comment about the first BP (Best Practice x: Provide
> Versioning Information) that you wrote.
> I would like to suggest to change the possible approach for
> implementation, where you have suggested to use numbers for version, to use
> SchemaVersion [1] format for version numbers.
>
> I’m writing a BP where I recommend to use the SchemaVer.
>
> What do you think about this?
>
> Newton
>
> PS: Do you use some kind of instant messaging, such as GTalk or Skype,
> because it would be easier and faster to talk about some small changes
> using it.
> I use both (newtoncalegari@gmail.com and Skype: newtoncalegari), if some
> of you want to add me to make easier to talk, feel free to do that.
>
>
> Em 14/01/2015, à(s) 17:22, Annette Greiner <amgreiner@lbl.gov> escreveu:
>
> Here is a start for the versioning best practices, enclosed in a Word file.
> -Annette
> <DataVersionBP.docx>
> On Jan 9, 2015, at 7:05 AM, Christophe Guéret <
> christophe.gueret@dans.knaw.nl> wrote:
>
> Hoi,
>
> If we need a use-case for versioning I could contribute one about
> classification schemes. I've been working a bit with the people from of the
> UDC on this issue. So far they produced http://udcdata.info/ which is a
> Linked Open Data export on of one version of the UDC summary. They are now
> looking into exporting the entire UDC scheme and all its past 17 revisions
> (one per year) that are digitally available right now. The challenge there
> is that users must have a version-less URI to point to. They should also be
> given the opportunity to point to a specific revision of a given resource
> when needed. They also need to be able to mint URIs that don't exist and
> get them to de-reference but that's a different issue... :-)
>
> So far we looked into taking advantage of the split between unique
> identifiers for resources and identifiers for their descriptions + the
> redirect mechanism to implement something such as "eg:id/22" that points to
> "eg:data/MRF10/22" with a "latest by default policy" or "eg:data/MRF02/22"
> when a query such as "eg:22?rev=02" is received. To let users point to
> specific time-stamped resources without having to use a GET parameter we
> thought about replacing the "id" by the version number "eg:02/22". For this
> to work the URI scheme of the resources would include a path element that
> can be replaced by a timestamp, something to be described as part of the
> metadata of the dataset (in this case the "id" to be replaced by the
> version).
>
> Christophe
>
>
>
> On 8 January 2015 at 18:38, Annette Greiner <amgreiner@lbl.gov> wrote:
> Writing introductions before content feels backward to me. It’s generally
> much easier to write an intro once you know what the content will be, and
> then you don’t have to worry about whether the assumptions you made still
> hold true.
> -Annette
> --
> Annette Greiner
> NERSC Data and Analytics Services
> Lawrence Berkeley National Laboratory
> 510-495-2935
>
> On Jan 8, 2015, at 9:24 AM, Mcgibbney, Lewis J (398M) <
> Lewis.J.Mcgibbney@jpl.nasa.gov> wrote:
>
> Hi Netwon,
>
> Anette, thank you for sharing your thoughts.
> +1
>
>
> I’m wondering how can we achieve the end of this subtask.
>
> So am I :)
>
>
> So, what I suggest for us is to get the Requirements related to Versioning
> topic.
> Maybe these two UCs [2][3] from the 2nd Round could be useful for this.
>
> It is my opinion that both of these use cases seem very worthy of our
> focus within the scope of this effort.
>
>
> Lewis, about the scope, I guess our first goal is to have a
> description/introduction of the Data Versioning section.
>
> Can you point me to the placeholder for this section? I am failing to find
> it.
>
> And we can discuss on tomorrow call about the BPs, because we don’t have
> yet requirements for Data Versioning to propose new ones.
>
> What do you think about it?
>
>
> Yes it sounds great
>
>
>
>
>
> --
> Onderzoeker
> +31(0)6 14576494
> christophe.gueret@dans.knaw.nl
>
> Data Archiving and Networked Services (DANS)
> DANS bevordert duurzame toegang tot digitale onderzoeksgegevens. Kijk op
> www.dans.knaw.nl voor meer informatie. DANS is een instituut van KNAW en
> NWO.
>
> Let op, per 1 januari hebben we een nieuw adres:
> DANS | Anna van Saksenlaan 51 | 2593 HW Den Haag | Postbus 93067 | 2509 AB
> Den Haag | +31 70 349 44 50 | info@dans.knaw.nl | www.dans.knaw.nl
>
> Let's build a World Wide Semantic Web!
> http://worldwidesemanticweb.org/
>
> e-Humanities Group (KNAW)
>
>
>
>
>
>
>
>
>


-- 
Onderzoeker
+31(0)6 14576494
christophe.gueret@dans.knaw.nl

*Data Archiving and Networked Services (DANS)*

DANS bevordert duurzame toegang tot digitale onderzoeksgegevens. Kijk op
www.dans.knaw.nl voor meer informatie. DANS is een instituut van KNAW en
NWO.


Let op, per 1 januari hebben we een nieuw adres:

DANS | Anna van Saksenlaan 51 | 2593 HW Den Haag | Postbus 93067 | 2509 AB
Den Haag | +31 70 349 44 50 | info@dans.knaw.nl <info@dans.kn> |
www.dans.knaw.nl


*Let's build a World Wide Semantic Web!*
http://worldwidesemanticweb.org/

*e-Humanities Group (KNAW)*
[image: eHumanities] <http://www.ehumanities.nl/>