The synopsis drawer

Thomas Schilling nominolo at googlemail.com
Tue Aug 3 08:22:41 EDT 2010


On 2 August 2010 13:44, Mark Lentczner <markl at glyphic.com> wrote:
> [I e-mailed this to the haddock mailing list last night - but as of this morning, it still hasn't been resent. Hence, I'm resending, and addressing a few of you directly. Sorry for any duplicate messages that arise.]
>
> ----
>
> I have built a new look for the synopsis drawer, and changed it from being a "roll over to open" to being a click on tab to open:
>
>        http://www.ozonehouse.com/mark/snap-xhtml/snap-core/Snap-Types.html
>
> You can click the tab again to close it, or just picking an item both scrolls to the item and closes it.
>
> Some people suggested we try making the synopsis more like the "mini_" pages. I have a rough version of that here (note that the links in this version don't work, it is just for evaluating the look):
>
>        http://www.ozonehouse.com/mark/snap-xhtml/snap-core-alt/Snap-Types.html
>
> For the record, some discussion with various people on IRC last week netted at least one strong opinion that the type signatures in the synopsis (as in the first version) are very helpful in knowing what's what.

Seeing them in direct comparison, I now tend to agree that it's a lot
less helpful without type signatures.  I wonder whether wrapping +
indenting the type signatures would help.  I.e., margin-left: 1.5em;
text-indent: -1em.  I'm not too convinced that this could work,
though.

The Scala docs by default show a more synopsis-like view.  They only
show one line of the description and clicking on it reveals the rest.
I don't really like it though; their default font is tiny, there's no
indication whether there's more documentation available or not, and
because you have to click on stuff it's not good for browsing.

The drawer looks fine -- I do not feel compelled to click, but it is
obvious that it is clickable.  The arrows should probably point the
other way around once it's expanded, but otherwise it's fine.

I don't have any concrete ideas at the moment.  I would have to
experiment myself.

/ Thomas



More information about the Haddock mailing list