[rfc-i] Should figures float?

FWIW, 'float' is a lot more complex than it seems. The float is often 
tethered, i.e., there are constraints as to how far away from the 
original section the float can occur. Typically it has to stay within 
the smallest numbered subsection and can't precede it's first reference 
unless on the same page.

I don't know whether any of that can be supported automatically within XML.

Joe

On 7/10/2014 11:53 PM, Riccardo Bernardini wrote:
> On Fri, Jul 11, 2014 at 8:22 AM, Julian Reschke <julian.reschke at gmx.de> wrote:
>> On 2014-07-11 03:29, Brian E Carpenter wrote:
>>>
>>> While reviewing the PDF draft, it occurred to me that most "modern"
>>> solutions (at least back to LaTeX) allow figures to float, whereas
>>> the assumption with XML2RFC has always been that the occur exactly
>>> where <figure> is placed in the text, except for maybe inserting a page
>>> feed to avoid splitting the artwork.
>>>
>>> For v3, are we going to allow for processors that are capable of
>>> floating a figure? That would be a new attribute for <figure>
>>> along the lines of float="true" or "false".
>>
>>
>> I believe it's an interesting idea, but on the long list of things we still
>> need to do it should be at the bottom.
>>
>> Also, I suspect that a simple flag isn't sufficient. Depending on whether
>> the image (what about tables???) floats, we may need different text content.
>
> My opinion in brief: definitively a +1 for having figures (and tables,
> and [possibly] algorithms, and...) floating for paginated formats: the
> result looks usually order of magnitude better than the result with
> fixed position.   Should I take a decision, I would go for floating by
> default + possibility of some adjustment via optional attributes (yes,
> it is the LaTeX model... :-).
> My personal experience  (> 20 years long) in writing scientific paper
> is that in the 99% of the cases the automatic placement algorithm
> makes a very good work, in the remaining 1% you may need some hand
> tweaking (done at the very last moment, in our case before finalizing
> the RFC).
>
> I'm not sure if floating would be useful in non-paginated format.  I
> would keep it, nevertheless, unless it turns out too much difficult to
> implement.
>
> Best regards,
>
> Riccardo
>
>>
>> Best regards, Julian
>>
>>
>> _______________________________________________
>> rfc-interest mailing list
>> rfc-interest at rfc-editor.org
>> https://www.rfc-editor.org/mailman/listinfo/rfc-interest
> _______________________________________________
> rfc-interest mailing list
> rfc-interest at rfc-editor.org
> https://www.rfc-editor.org/mailman/listinfo/rfc-interest
>