What's the determining factor in the UI for pdfs on the web?
OTOH it must be the server. For pdfs served up by an archive
eg. 

http://web.archive.org/web/20160404161549/https://code.google.com/p/xpost/downloads/detail?name=spiral-fill.pdf&can=2&q=

You have to just download it. But for other pdfs, served up
as part of a "document-service" (for want of more informed
terminology) can feed a bit of needed pre-parsed metadata
that the browser plug-in can chug with. eg.

http://arxiv.org/pdf/1605.06640v1.pdf

And for a specially-powered few of these "document service"
httpd extension suites, it can feed the xref tables first
(which lie physically at the *end* of the pdf file) enabling
the plugin to start displaying page 1 as soon as its elements
are present. eg.

http://algorithmicbotany.org/papers/abop/abop.lowquality.pdf

And still-more-powerful systems like GitHub can even wrap
their own html around it. eg.

https://github.com/luser-dr00g/ibis.ps/blob/master/ibis.pdf

Am I right? Apache packages?

On Tue, 21 Jun 2016 22:02:03 -0700, luserdroog wrote:

> What's the determining factor in the UI for pdfs on the web?
> OTOH it must be the server. For pdfs served up by an archive eg.
> 
> http://web.archive.org/web/20160404161549/https://code.google.com/p/
xpost/downloads/detail?name=spiral-fill.pdf&can=2&q=
> 
> You have to just download it. But for other pdfs, served up as part of a
> "document-service" (for want of more informed terminology) can feed a
> bit of needed pre-parsed metadata that the browser plug-in can chug
> with. eg.
> 
> http://arxiv.org/pdf/1605.06640v1.pdf
> 
> And for a specially-powered few of these "document service"
> httpd extension suites, it can feed the xref tables first (which lie
> physically at the *end* of the pdf file) enabling the plugin to start
> displaying page 1 as soon as its elements are present. eg.
> 
> http://algorithmicbotany.org/papers/abop/abop.lowquality.pdf
> 
> And still-more-powerful systems like GitHub can even wrap their own html
> around it. eg.
> 
> https://github.com/luser-dr00g/ibis.ps/blob/master/ibis.pdf
> 
> Am I right? Apache packages?

PDF file has to be optimized for web delivery (can be done when creating
or later by utilities.) And the web server has to support "ranges" aka 
byte serving.

> What's the determining factor in the UI for pdfs on the web?

I don't think you're looking at the right components.  But perhaps it
would help you to know that the pages in a PDF can be reordered
internally without changing the presentation order in a process called
"linearizing".  E.g. "qpdf --linearize <pdf file>" will ensure the
first page loads immediately.