Re: dwbp-ISSUE-206: The example with memento needs to be rewritten to use the transport example [Best practices document(s)]

Hi Herbert,

Thanks a lot for your explanation!  If I understood correctly, versioning
information is described through the TimeMap and the TimeGate.

In the current version of the examples, we show how to get information
about versioning through TimeMaps and TimeGates, but we don't show how to
"create" them (TimeMaps and TimeGates).

IMO, If we're gonna use Memento as an approach for implementation then I
think we should provide a short explanation of how to create TimeMaps and
TimeGates. We can also provide a link to a page with an explanation. Does
it make sense for you?

Cheers,
Bernadette



2015-11-06 10:15 GMT-03:00 Herbert Van de Sompel <hvdsomp@gmail.com>:

> hi Bernadette,
>
> I am not sure whether I understand your question. But let me try. If
> the below does not help, then I hope that
> http://mementoweb.org/guide/howto/ does.
>
> Assume:
> * http://example.org/dataset is the “generic URI” at which the current
> version of a dataset is always available
> * http://example.org/dataset-000 is the versioned URI of the first
> version of the dataset, which was "current" in the temporal interval
> [t1,t2[
> * http://example.org/dataset-001 is the versioned URI of the second
> version of the dataset, which was "current" in the temporal interval
> [t2,t3[
> * http://example.org/dataset-002 is the versioned URI for the current
> version of the dataset, which became current at t3 and still is now
>
> Under these assumptions, Memento provides the following functionality:
>
> => Datetime Negotiation functionality - Given
> http://example.org/dataset , follow a "timegate" typed link to a
> TimeGate. At the TimeGate, perform datetime negotiation using a time
> tq, with t0 <= tq <= now . This will lead to the version of the
> dataset that was current at tq. For example, if t1 <= tq < t2 the
> resulting URI will be http://example.org/dataset-000 . If the temporal
> dataset version is available in different serializations, then regular
> content negotiation for a specific serialization is also possible.
>
> => Version history functionality - Given http://example.org/dataset
> follow a "timemap" typed link to a TimeMap. Accessing the TimeMap will
> yield a document that includes the following info:
>
> http://example.org/dataset
> http://example.org/dataset-000; datetime="t1"
> http://example.org/dataset-001; datetime="t2"
> http://example.org/dataset-002; datetime="t3"
>
> If these temporal versions are available in different serializations,
> this document could include the following info:
>
> http://example.org/dataset
> http://example.org/dataset-000; datetime="t1"; type="type1"
> http://example.org/dataset-001; datetime="t2"; type="type1"
> http://example.org/dataset-002; datetime="t3"; type="type1"
> http://example.org/dataset-000; datetime="t1"; type="type2"
> http://example.org/dataset-001; datetime="t2"; type="type2"
> http://example.org/dataset-002; datetime="t3"; type="type2"
>
> => Generic URI functionality - Given the URI of a version, eg
> http://example.org/dataset-001 , follow a "original" typed link to the
> generic URI for the dataset, ie to http://example.org/dataset
>
> Please let me know whether this answered your question.
>
> Cheers
>
> Herbert
>
>
>
> On Fri, Nov 6, 2015 at 4:35 AM, Bernadette Farias Lóscio
> <bfl@cin.ufpe.br> wrote:
> > Hi Herbert,
> >
> > Thanks a lot for your message! We made the corrections on the example
> [1] .
> >
> > I have another question about the Memento protocol. You said that
> Memento is
> > a protocol that "Access the version of a resource that was the current
> > version at a specified date/time." My question is: when we have multiple
> > versions of a resource how memento  identifies which version will be the
> > current version at a specified time? I'm sorry if this is a stupid
> question,
> > but I'm trying to find a better way to connect memento with the BP.
> >
> > Cheers,
> > Bernadette
> >
> > [1] http://w3c.github.io/dwbp/bp.html#dataVersioning
> >
> >
> >
> > 2015-11-05 5:38 GMT-03:00 Herbert Van de Sompel <hvdsomp@gmail.com>:
> >>
> >> On Wed, Nov 4, 2015 at 8:13 AM, Bernadette Farias Lóscio
> >> <bfl@cin.ufpe.br> wrote:
> >> > Hi Herbert,
> >> >
> >> > I just updated the DWBP document to include the versioning example
> with
> >> > memento [1]. Thanks a lot for your contribution!
> >> >
> >> > Please, take a look and tell me if it is ok!
> >> >
> >>
> >> Three things:
> >>
> >> * The link to Memento info seems to have disappeared. The official
> >> spec is RFC7089, whereas http://mementoweb.org/howto/ has more on
> >> Memento and versioning.
> >>
> >> * In Example 8, it would be nice to have a blank line after each curl
> >> command.
> >>
> >> * Something seems to have gone wrong with the second curl in Example
> >> 9.  This is what it should look like:
> >>
> >> curl -i http://example.org/timemap/dataset
> >>
> >> HTTP/1.1 200 OK
> >> Content-Type: application/link-format
> >>
> >> <http://example.org/dataset>;rel="original”,
> >> <http://example.org/timegate/dataset>;rel="timegate”,
> >> <http://example.org/timemap/dataset>;rel="timemap”;
> >> type=“application/link-format”,
> >> <http://example.org/dataset-000>; rel=“first memento";
> >> datetime="Thu,05 Mar 2015 00:00:00 GMT",
> >> <http://example.org/dataset-001>; rel=“memento"; datetime=“Sun, 05
> >> Apr2015 00:00:00 GMT"
> >> <http://example.org/dataset-002>; rel=“last memento"; datetime="Tue,05
> >> May 2015 00:00:00 GMT"
> >>
> >> > I also have one question about the use of memento. If I understand
> >> > correct,
> >> > Memento is a protocol that helps to obtain different versions of the
> >> > same
> >> > resource.
> >>
> >> Memento (RFC7089) is an HTTP-based protocol that provides the
> >> following functionality regarding resource versions:
> >> * Access the version of a resource that was the current version at a
> >> specified date/time. This is achieved via datetime negotiation, a
> >> variant of content negotiation, and involves a so-called TimeGate.
> >> * Access the version history of a resource. This is achieved by
> >> obtaining a TimeMap, which lists all resource versions and their
> >> respective date/time.
> >>
> >> > In this case, I think we should add something on the BP to make
> >> > it clear that we need: i) a way to describe and to add versioning
> >> > information to datasets and ii) a way to access such information and
> to
> >> > access different versions of a resource. I was also wondering if
> instead
> >> > of
> >> > having this information as part of the BP08 and BP09, it would be
> better
> >> > to
> >> > have a specific BP related to provide access to different dataset
> >> > versions.
> >> > Does it make sense for you?
> >> >
> >>
> >> * I think that the Memento TimeMap example in BP09 aligns well with
> >> the title of the BP "Provide version history", which is exactly what
> >> TimeMaps do. However, it fits less with the description in the "Why"
> >> of that BP which states "Determining how the various versions of a
> >> dataset differ from each other is typically very laborious unless a
> >> summary of the differences is provided." Memento TimeMaps merely
> >> provide an overview of the URIs of resource versions and their
> >> creation datetimes. TimeMaps provide no info on what the difference
> >> between the versions is, other than that they were created at
> >> different times. With that regard, it fits better under BP08, which is
> >> about uniquely identifying versions, etc.
> >>
> >> * I think you are right to suggest that the Memento example in BP08 is
> >> not really about "Providing version information". With that regard, as
> >> per the above, the TimeMap example would seem to fit better under
> >> BP08. The current Memento example in BP08 is about accessing a version
> >> of a dataset that was the "current" version at some given date/time in
> >> the past. This feature, as such, is not formulated as a BP. Note that
> >> TimeMaps can also provide this feature since they list all versions.
> >> But, if there are a very significant number of versions, the use of
> >> TimeMaps for access to a specific temporal version can be cumbersome.
> >>
> >> Since I am at it, I would like to add some thoughts related to the
> >> "webby" stuff that Erik and I have brought up. I think it would be
> >> very constructive to include an additional Best Practice about using
> >> typed links as a way to guide clients and applications. In this new
> >> Best Practice, a few examples could be provided that show HTTP
> >> interactions/headers that provide typed web links. The following
> >> examples come to mind, immediately:
> >> - the describes/describedby links between dataset and metadata about
> >> dataset, which I mailed about previously
> >> - prev/next links to navigate between versions of datasets
> >> - the Memento timemap link for access to version history
> >> - maybe others, for example, one using the duplicate relation type to
> >> link to a mirror copy
> >>
> >> Personally, I think this would be a crucial best practice to add to
> >> the document.
> >>
> >> Cheers
> >>
> >> Herbert
> >>
> >>
> >> > Thanks,
> >> > Bernadette
> >> >
> >> > [1] http://w3c.github.io/dwbp/bp.html#dataVersioning
> >> >
> >> >
> >> >
> >> >
> >> >
> >> > 2015-10-23 13:42 GMT-03:00 Bernadette Farias Lóscio <bfl@cin.ufpe.br
> >:
> >> >>
> >> >> Hi Herbert,
> >> >>
> >> >> Thanks a lot for your help! It would be great if you could help us
> with
> >> >> the rewriting of the versioning examples.
> >> >>
> >> >> Cheers,
> >> >> Bernadette
> >> >>
> >> >>
> >> >>
> >> >> 2015-10-23 11:40 GMT-02:00 Herbert Van de Sompel <hvdsomp@gmail.com
> >:
> >> >>>
> >> >>> hi all,
> >> >>>
> >> >>> => I am willing to provide material to rewrite the existing Memento
> >> >>> example to match the example that is used throughout the document.
> >> >>>
> >> >>> => I think that Memento should also be mentioned in the version
> >> >>> history section. Memento has the notion of TimeMaps, documents that
> >> >>> list the entire temporal version history of a resource. See
> >> >>> http://mementoweb.org/guide/rfc/#Pattern6-Response1 . I am willing
> to
> >> >>> provide that material too.
> >> >>>
> >> >>> Pls let me know if OK.
> >> >>>
> >> >>> Cheers
> >> >>>
> >> >>> Herbert
> >> >>>
> >> >>> On Thu, Oct 22, 2015 at 4:54 PM, Data on the Web Best Practices
> >> >>> Working Group Issue Tracker <sysbot+tracker@w3.org> wrote:
> >> >>> > dwbp-ISSUE-206: The example with memento needs to be rewritten to
> >> >>> > use
> >> >>> > the transport example [Best practices document(s)]
> >> >>> >
> >> >>> > http://www.w3.org/2013/dwbp/track/issues/206
> >> >>> >
> >> >>> > Raised by: Bernadette Farias Loscio
> >> >>> > On product: Best practices document(s)
> >> >>> >
> >> >>> > The example used in BP - Information about dataset versioning
> should
> >> >>> > be
> >> >>> > available is not consistent with other examples used in the
> >> >>> > document.
> >> >>> >
> >> >>> >
> >> >>> >
> >> >>>
> >> >>>
> >> >>>
> >> >>> --
> >> >>> Herbert Van de Sompel
> >> >>> Digital Library Research & Prototyping
> >> >>> Los Alamos National Laboratory, Research Library
> >> >>> http://public.lanl.gov/herbertv/
> >> >>>
> >> >>> ==
> >> >>>
> >> >>
> >> >>
> >> >>
> >> >> --
> >> >> Bernadette Farias Lóscio
> >> >> Centro de Informática
> >> >> Universidade Federal de Pernambuco - UFPE, Brazil
> >> >>
> >> >>
> >> >>
> ----------------------------------------------------------------------------
> >> >
> >> >
> >> >
> >> >
> >> > --
> >> > Bernadette Farias Lóscio
> >> > Centro de Informática
> >> > Universidade Federal de Pernambuco - UFPE, Brazil
> >> >
> >> >
> ----------------------------------------------------------------------------
> >>
> >>
> >>
> >> --
> >> Herbert Van de Sompel
> >> Digital Library Research & Prototyping
> >> Los Alamos National Laboratory, Research Library
> >> http://public.lanl.gov/herbertv/
> >>
> >> ==
> >
> >
> >
> >
> > --
> > Bernadette Farias Lóscio
> > Centro de Informática
> > Universidade Federal de Pernambuco - UFPE, Brazil
> >
> ----------------------------------------------------------------------------
>
>
>
> --
> Herbert Van de Sompel
> Digital Library Research & Prototyping
> Los Alamos National Laboratory, Research Library
> http://public.lanl.gov/herbertv/
>
> ==
>



-- 
Bernadette Farias Lóscio
Centro de Informática
Universidade Federal de Pernambuco - UFPE, Brazil
----------------------------------------------------------------------------