[openspending-dev] OpenSpending - questions & suggestions

[Friedrich Lindenberg]

Hi,

let me try and go through these quickly:

On Fri, Aug 3, 2012 at 10:21 PM, račan.sk <racan.sk at gmail.com> wrote:
> Questions:
> - why some pages (forms) need manual refresh while others do not?

Can you give more details please: which ones?

> - why are the visualizations not displayed correctly in Google Chrome?
> (overlapping buttons and/or forms and the bottom of the page)

Same here :)

> - can loading data into a dataset be safely stopped and cancelled?

Well, you cannot stop it at all since it is independent of the web
interface. This is also why you need to refresh the web interface to
see how many entries have already been loaded.

> - on hovering over the bubbles in the BubbleTree visualization, a message
> "No description given" is shown. Where can the description be entered?

In the dimension that the bubble represents, add an attribute called
"description" for each entry in the data.

> Suggestions:
>
> Settings:
> - customization of the number of records displayed in one page (browsing
> form as well as visualizations)

Hm, I'm one of the people who think that we should rather invest a bit
of time to find out what the right number of results to show is,
rather than making such things an option (that's how you end up with
MS Excel - 10000 buttons but no clear use).

> Visualizations:
> - the 'Data' button exports only the displayed data - sometimes it could be
> useful to export the complete dataset.

If you want to download the dataset, you can go to the "About" tab on
the dataset page and click the link pointing to the sources.
Alternatively, there is a way to get the data out of OpenSpendings
APIs - but since it makes a lot of server load, we're not linking to
it anywhere. For most people, downloading a 500 MB source file is just
less useful than downloading the concrete table they're looking at.

> - Amount column: right align could be more appropriate for the numeric
> values (generalization: defining alignment of every field in the data model)

Aye.

> - labels defined in the data model could be used as column headings

They are?

> Dimensions & Measures Tab:
> - label "Exisiting Dimensions" should be "Existing Dimensions"
> - as a user, I'd prefer user-defined order of "Existing Dimensions", thus
> giving users the opportunity to order them more logically than by sorting
> them alphabetically
> - in the list of dimensions, labels would seem to me more appropriate (the
> special types from, to, time, amount in parentheses could be in parenteses)

That's right, good point.

> Automatic Dataset update:
> - any time a newer CSV source file is localized by the OpenSpending portal,
> the dataset is automatically deleted, the newer data loaded and an update
> report link sent to the owner's e-mail address

There is no need to delete any data, OS is built to support in-place
updates so that only new rows will be added to the dataset.

> Paging buttons:
> - very impractical in case of datasets with higher number of pages; is it
> not possible to use buttons 'first' and 'previous', then text box 'page
> number', and then buttons 'next' and 'last'?

Aye, good point.

Cheers,

 - Friedrich