CDR Tickets

BZDATETIME::2011-09-01 12:20:49
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::1

Adding email summarizing meeting with Bryan. Additional notes from Bob are inserted below:

From: Beckwith, Margaret (NIH/NCI) [E]
Sent: Monday, August 29, 2011 4:44 PM
To: Kline, Robert (NCI); Pizzillo, Bryan (NIH/NCI) [C]
Subject: mobile attribute in cdr

Hi Bob and Bryan,

I was looking at my notes from the Aug. 8 CDR requirements meeting we had to discuss mobile impact on CDR, and wanted to make sure I understood what we agreed on. So here is a summary from my notes:

Exclude/Include attributes
• We will add a pair of para-level and section-level attributes called “excluded device” and “included device”. The attribute list of values will be screen, mobile, ebook for now. Note from BOB: We'll probably go with "ExcludedDevices" and "IncludedDevices"
to conform to the style used in similar multi-word attribute names in the DTD (and to reflect the fact that the attribute can identify more than one device type).

• The addition of one of these attributes on a para-level element or section will trump the default to publish to every device.

• It will be possible to nest an “exclude” but not an “include” attribute. Note from Bob: The rest looks OK, but I think what we decided for nesting was that the only thing you weren't allowed to do was to turn back on publishing portion "B" of a document to a device if "B" is nested inside "A" and publishing for that device had been turned off for "A." So

<A IncludeDevices="mobile ebook">
:
<B IncludeDevices="mobile">...</B>
</A>

is OK (further restricting the devices for which B is included), but

<A IncludeDevices="mobile">
:
<B IncludeDevices="mobile screen">...</B> </A>

is not allowed. In other words, you can drop publishing targets
("devices") as you dig further in the nesting, but you can't go the other way (adding devices back in). You could come up with other examples which mix IncludeDevices and ExcludeDevices at different levels of nesting, and by applying that same principle determine which combinations are allowed and which are not.

• We will programmatically add “excluded device” = mobile to those sections of the patient summaries that we do not want to appear in the mobile cancer.gov release.

Mobile pretty url
• We will add an element for a mobile pretty url in the metadata section of the summaries (where the regular pretty url is).

• Lisa will review the site map in Percussion and will give us a spreadsheet with urls to populate the element in the CDR.

Please feel free to correct if I got something wrong, or if the summary is incomplete. After we agree on this, I will put the comments into the issue in bugzilla.

Thanks a lot,

Margaret

BZDATETIME::2011-09-06 09:05:18
BZCOMMENTOR::Bob Kline
BZCOMMENT::2

Do we want to go ahead and modify the schemas now, or hold off on that since it was decided that the first use of this mechanism would be injected during publishing?

BZDATETIME::2011-09-08 10:18:07
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::3

Let's just do it through publishing for now and we can modify the schema when we need it.

BZDATETIME::2011-09-12 09:24:24
BZCOMMENTOR::Bob Kline
BZCOMMENT::4

(In reply to comment #3)
> Let's just do it through publishing for now and we can modify the schema when
> we need it.

I believe you said you had second thoughts about this decision, Margaret. If so, for which of the following elements do you want to have the attributes available?

Para and LiteralLayout (same complex type)
Table
ItemizedList
OrderedList
QandASet
Contact
MediaLink
MiscellaneousDocLink
PlaceHolder
SummaryModuleLink
SummarySection

BZDATETIME::2011-09-15 14:34:30
BZCOMMENTOR::Bob Kline
BZCOMMENT::5

(In reply to comment #4)

> so, for which of the following elements do you want to have the attributes
> available?

All of them.

BZDATETIME::2011-09-30 16:17:21
BZCOMMENTOR::Bob Kline
BZCOMMENT::6

IncludedDevices and ExcludedDevices attributes installed on Mahler; ready for user testing.

BZDATETIME::2011-10-06 11:25:45
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::7

(In reply to comment #6)
> IncludedDevices and ExcludedDevices attributes installed on Mahler; ready for
> user testing.

There seems to be a problem with the Summary Schema on Mahler. Summary documents are coming up with Rules checking errors. Most of them have to do with the PDQBoard element.

BZDATETIME::2011-10-26 12:08:28
BZCOMMENTOR::Bob Kline
BZCOMMENT::8

(In reply to comment #7)
> (In reply to comment #6)
> > IncludedDevices and ExcludedDevices attributes installed on Mahler; ready for
> > user testing.
>
> There seems to be a problem with the Summary Schema on Mahler. Summary
> documents are coming up with Rules checking errors. Most of them have to do
> with the PDQBoard element.

I made the MobileURL element optional (as we did on Bach).

BZDATETIME::2011-11-08 12:07:21
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::9

A couple of comments:

1. The mobile cancer.gov Web site will include the PDQ Adult AND Pediatric treatment summaries for patients (not just Adult as it says in the summary).

2. I created a test document on Mahler that has the "exclude from mobile" attribute on the image table, and the "include in mobile" attribute on two media links for the individual images. I can add the mobile url element whenever I get the information from Bryan.

BZDATETIME::2011-11-08 13:19:00
BZCOMMENTOR::Volker Englisch
BZCOMMENT::10

A comment (or more a question) from me as well.

How would you enter to exclude a section from two devices? The drop down list only allows you to enter a single device.

BZDATETIME::2011-11-08 13:25:36
BZCOMMENTOR::Bob Kline
BZCOMMENT::11

I updated the schema to allow all the possible combinations of values for the attribute.

BZDATETIME::2011-11-08 14:02:51
BZCOMMENTOR::Volker Englisch
BZCOMMENT::12

There is valuable information for the vendor filter changes in this issue (comment #9), so I'll mark my and this issue as dependent.

BZDATETIME::2011-11-15 10:35:57
BZCOMMENTOR::Volker Englisch
BZCOMMENT::13

(In reply to comment #11)
> I updated the schema to allow all the possible combinations of values for the
> attribute.

I've tried to enter 'mobile screen' as one combination but the validation failed.

BZDATETIME::2011-11-15 15:29:43
BZCOMMENTOR::Bob Kline
BZCOMMENT::14

(In reply to comment #13)

> I've tried to enter 'mobile screen' as one combination but the validation
> failed.

Hmm. I put 'screen mobile' in the schema for that combination. Let's discuss the best solution in Thursday's meeting.

BZDATETIME::2011-11-17 13:24:37
BZCOMMENTOR::Bob Kline
BZCOMMENT::15

Just put screen and mobile in the valid values list for now.

BZDATETIME::2011-11-17 13:50:30
BZCOMMENTOR::Bob Kline
BZCOMMENT::16

(In reply to comment #15)
> Just put screen and mobile in the valid values list for now.

Also add "none" for ExcludedDevices; this won't be exported as such by the publishing filters, but will instead be interpreted by the filter as an instruction not to add "ExcludedDevices='mobile'" to an element which would normally get that treatment. Instead the ExcludedDevices="none" attribute is stripped by the filter.

BZDATETIME::2011-11-22 08:27:19
BZCOMMENTOR::Bob Kline
BZCOMMENT::17

Latest changes made to schema on Mahler; ready for user review and testing.

BZDATETIME::2011-12-08 07:35:14
BZCOMMENTOR::Bob Kline
BZCOMMENT::18

[This is pasted in from a thread I picked up from the CG Percussion team; the thread is backwards, so the older messages are after the newer ones. We should discuss the questions in this afternoon's CDR status meeting.]

Pizzillo, Bryan (NIH/NCI) [C] bryanp@mail.nih.gov
2:48 PM (16 hours ago)

to me, Margaret, Blair, Alan, Volker, Ron
So, we may be able to suppress the link; however, they don't usually make sure that wording does not imply a link.

---Original Message---
From: Kline, Robert (NCI)
Sent: Wednesday, December 07, 2011 2:38 PM
To: Pizzillo, Bryan (NIH/NCI) [C]; Beckwith, Margaret (NIH/NCI) [E]
Cc: Learn, Blair (NIH/NCI) [C]; Alan Meyer; Englisch, Volker (NIH/NCI) [C]; Burr, Ron (NIH/NCI) [C]
Subject: Re: SummaryRef rules

I'll bring these questions up at tomorrow's CDR status meeting. It's
possible the users were hoping that CG would detect that a link target
was only available on platforms other than the one used by the current
visitor and suppress the <a/> markup (of course, it would be the
editor's responsibility to avoid wording which would imply that a link
were present). It's also possible that they haven't thought about the
questions (I hadn't).

Bob

On Wed, Dec 7, 2011 at 1:29 PM, Pizzillo, Bryan (NIH/NCI) [C]
<bryanp@mail.nih.gov> wrote:
> Comments between [] for those text email users. When thinking about the
> linking remember that we want to keep users on the mobile site if at all
> possible. Once they are transferred to the desktop from mobile there is no
> way navigating back to mobile but through their back button or the mobile
> home page link at the top of all desktop pages.
>
> From: Learn, Blair (NIH/NCI) [C]
> Sent: Wednesday, December 07, 2011 1:02 PM
> To: Kline, Robert (NCI); 'Alan Meyer'; Englisch, Volker (NIH/NCI) [C];
> Pizzillo, Bryan (NIH/NCI) [C]; Burr, Ron (NIH/NCI) [C]
> Subject: SummaryRef rules
>
> In our discussions of moving summaries to mobile, I don't recall that we
> discussed SummaryRef elements in much detail. I'm working on how to resolve
> them on mobile, and need to double-check some overall assumptions. I
> anticipate the second set of bullets is the one which will generate the most
> conversation.
>
> Here's how desktop works. If these assumptions are incorrect, then we
> already have problems.
>
> · A reference to a section within the document will always resolve
> to that document. (e.g. within summary 62995, if there's a reference to
> 62995#_1234, then section _1234 will definitely exist.)
>
> · A reference to another summary will always resolve. (e.g. within
> summary 62995, if there's a reference to 62787, then summary 62787 will
> definitely exist.)
>
> · A reference to another a section within another summary will
> always exist. (e.g. within summary 62995, if there's a reference to
> 62787#_1234, then summary 62787 will definitely exist and will definitely
> contain a section _1234.)
>
> Where I'm uncertain is how things work once we start turning sections on and
> off and don't include all the summaries. Are these assumptions true?
>
> · A reference to a section within the document will always resolve
> to that document. (e.g. within summary 62995, if there's a reference to
> 62995#_1234, then section _1234 will definitely exist.)
>
> o My key assumption here is that if a section is not available on a given
> device, then it's also invalid to reference it from a section which is
> included.
>
> [Could there be a rule in the CDR that stops users from linking to (or
> publishing) another section where the target does not have the same level of
> visibility as the source? Basically, a link from a section with all
> visibilities -> mobile only section should fail; a mobile section to a
> mobile section would be allowed.]
>
> · A reference to another summary will always resolve, but perhaps
> not on the same site. (e.g. If the mobile version of 62995 references
> summary 62787, then it's possible that summary 62787 will only exist on the
> desktop site.)
>
> [That is fine, this will most likely happen with links to screening and CAM
> summaries. Just make the URL /whatever and Lee's unmanaged handler should
> handle it. We should check that anchor is kept with the redirect and that
> this is not a very bad idea. ]
>
> · A reference to another section in another summary will always
> resolve, but perhaps not on the same site. (e.g. If the mobile version of
> 62995 references 62787_1234, if summary 62787 only exists on the desktop
> site, it will contain section _1234.)
>
> [If implemented the check two bullet points up should handle that]
>
> · Not sure how to phrase this one. If a section is referenced in a
> separate summary, and the separate summary exists on multiple devices, then
> the section will exist on the same device as the reference.
>
> o So for example, if summary 62995 references 62697#_1234, and both
> summaries exist on desktop and mobile, and the reference appears on mobile,
> it will also resolve on mobile.
>
> [I would say that is correct as well.]

BZDATETIME::2011-12-08 16:54:39
BZCOMMENTOR::Bob Kline
BZCOMMENT::19

I met with Blair this afternoon, and relayed the users' response that they will almost always be linking to a target which is present on at least all of the devices on which the link itself appears. In the rare case when they link to a target which is not present on all of the devices on which the link appears they will be careful to make sure the wording stands on its own without the link markup. Blair said he will need to do some further investigation about whether it will actually be possible to suppress the link markup in such cases.

BZDATETIME::2012-01-27 11:48:40
BZCOMMENTOR::Bob Kline
BZCOMMENT::20

See http://verdi.nci.nih.gov/tracker/show_bug.cgi?id=5089#c18

BZDATETIME::2012-01-30 11:53:13
BZCOMMENTOR::Bob Kline
BZCOMMENT::21

Schemas (CdrCommonBase and SummaryCommon) promoted to Bach Friday night.

BZDATETIME::2012-02-16 11:57:30
BZCOMMENTOR::Margaret Beckwith
BZCOMMENT::22

Closing issues since mobile site launched without any issues.