[humanities-dev] New textus version deployed

Tue Jun 26 12:25:47 UTC 2012

This looks fantastic Tom!

A few comments:

  * Generally this works well - apart from a few minor bugs which I'm
sure you're aware of (such as list of texts sometimes not loading
properly, or loading very slowly). So well done!

  * Feel like the UI/design has room for improvement. This should be
beautiful, and make you *want* to stay on the site and read. There are
loads of things to look up to in this regard, but my five top picks
would probably be:

    (i) substance.io
    (ii) Internet Archive's book reader, e.g.
http://publicdomainreview.org/2012/06/21/the-first-six-books-of-the-elements-of-euclid-1847/
    (iii) davidhume.org, a nice example of presenting philosophy
texts, simple interface, renders texts very beautifully:
http://www.davidhume.org/texts/thn
    (iv) Nietzsche Source - like davidhume.org this uses page metaphor
of white box for text, around which is grey / darker coloured
background: http://www.nietzschesource.org/#eKGWB/JGB-233
    (v) The White Review - http://www.thewhitereview.org/ - a site
which *may* have been inspired by the Public Domain Review in its
layout (it was designed by a friend of mine who knew and liked the
PDR), but which typography looks very nice

  * Regarding font/typography, feel like key ingredients would be:
bigger default text, serif font (as this is standard for printed
humanities texts), slightly thinner column - as the moment one risks
losing one's line. Font on davidhume.org is nice, not sure about
compatibility. Also White Review font is good.

  * Functionality / UI wishlist:

     - Toggle between single page view and scroll (so, e.g. I could do
a plain text search in my browser if I wanted to, or scroll to find
bit of text without having to click back and forth lots - the
equivalent of flicking through a book at speed)
    - Toggle annotations on and off - by default I'd like clean texts
first, and annotations as an afterthought/supplement (e.g. if I don't
know what something is, don't recognise an obscure word, want
contextual information about something I'm not familiar with). With
multiple annotations being shown, possible to toggle on/off by
clicking on annotation?
    - Toggle way to ambiently display the fact that a piece of text is
annotated, without distinguishing between who made them (e.g. very
light grey overlay). Kindle does this I think.
    - The 'page forward' and 'page backwards' buttons I feel should
naturally be strips either side of the page in the middle of the page,
in addition to at the bottom of the page. This would also make it
possible to use on a smart phone, which would be nice (at the moment
it is hard to hit buttons at the bottom).
    - Possible to have chapter forward, chapter back, and/or
forward/back 5/10 pages?
    - Options to change font size/style. Possible for users to choose
their own in settings when logged in?
    - With markup, do we have equivalent of title, h1, h2, etc.?
Again, davidhume.org does this beautifully with the "IM Fell English"
Google Web Font.
    - Not sure about current colours, bevels and appearance of
menus/interface. What about switching to standard Bootstrap top bar
for now, which - although perhaps not the most exiting in the world -
is thin, sleek and nondescript? I feel like this part of the design
should basically be invisible to the user, and text display should be
the centre of attention. Substance.io also uses something which is
similar in size, appearance to Bootstrap for their top bar(s). In
future, perhaps this is something that the user could customise
(basically pick their own admin interface colours, like Wordpress).
    - It is really important for users to be able to have some basic
information about the texts. At the moment none of the translations of
Plato list the translator as far as I can see. Ideally we'd have
information about the specific edition that the electronic text is
taken from, as well as the intermediary source (including URL) which
might be Wikisource, Project Gutenberg, or elsewhere on the web.
Important to have a way to track this provenance information. Also
useful way to navigate to other related resources.
    - As we can (in principle) have a URL to point to a specific piece
of text, is there any way we could enable user to select a piece of
text and get a URL?
   - Minor aside: I get special character errors with this text:
http://beta.openphilosophy.org/#text/zkl8mkycROyoSN_qyDgj6w/0

  * Regarding bibliography pages:

    - On the list of texts I think it should be possible to click on a
item to see the bibliographic metadata for that item, and also to
click 'READ' to go straight to the text view. This is important as we
also want to be able to have bibliographic items for which we do not
yet have full texts uploaded - so we can see what needs to be done.
    - Ideally it would be good to be able to tell from items on the
list view: (i) whether or not there is an associated text that you can
read in TEXTUS, (ii) whether or not there are associated URLs for
plain texts (not on TEXTUS, e.g. not yet extracted/marked
up/uploaded), (iii) whether or not there are scanned images available.
    - While I understand we won't be able to directly support
transcription in this version of the project, it would be good to give
users some sense of how they can get scans (e.g. from Internet
Archive), upload them somewhere (e.g. Wikisource) and get started on a
transcription. Also would be nice to be able to point to
transcriptions in progress as 'resources' attached to a given
bibliographic item. Sam and I could help to investigate this and
provide some documentation. If this is there then users can *do*
something if a text is missing and they really want to make sure it
gets added to TEXTUS. A killer feature would be to enable some
workflow for getting sections of a text up on PyBossa to enable
crowdsourced transcription, and to feature a selection of these on the
front page of OpenPhilosophy.org. ;-)
    - Not sure whether we're planning to support arbitrary
bibliographies (undergraduate reading lists, bibliographies for PhDs,
articles or monographs), but would be good to have option to add texts
to lists from text view and from bibliographic item view. Also would
like to export bibliographies in a variety of formats - e.g. for use
in Zotero, offline bibliography software, to embed in document in word
processor, etc.
    - Again - not sure of the status of this, but would be great to be
able to export texts in a variety of formats (PDF, epub, mobi, etc).

  * On upload/adding texts:

    - Rufus wrote a nice script to import texts from Project
Gutenberg, stripping out all header/footer text. Wonder if this could
be useful, e.g. to import via URL as well as upload?
    - Would be good to have a small link to a page that gives more
detail about markdown syntax and uploading texts.

I'll stop now. This email is already much bigger than I intended!

Bottom line: excited about progress and can't wait to watch this
develop further. :-)

All the best,

Jonathan

On Mon, Jun 25, 2012 at 4:14 AM, Tom Oinn <tom.oinn at okfn.org> wrote:
> Hi all,
>
> There's a new textus build at http://beta.openphilosophy.org - data
> store has been wiped so if you played with it before you'll need to
> re-register. The new build is a halfway house towards better support
> for bibliographic references (see recent blog here -
> http://blog.okfn.org/2012/06/20/bibliographic-references-in-textus/).
> In this build you can
>
> - review uploads before storing them (primarily to check whether the
> markup is being correctly interpreted)
> - assign proper bibliographic information to uploaded texts when reviewing
> - search text (bibliographic) metadata through the embedded facetview
>
> - embed this text search in your own site should you wish (left as an
> exercise for the reader - the 'bibserver-like' elasticsearch endpoint
> is at http://beta.openphilosophy.org/api/texts-es?, point the okfn
> facetview at this and you should see some texts. That or the server
> will explode. Please don't make the server explode...)
>
> This iteration does not include reading list functionality but almost
> all the groundwork for that is now done so it should appear soon[tm]!
>
> --
> Tom Oinn
> +44 (0) 20 8123 5142 or Skype ID 'tomoinn'
> http://www.crypticsquid.com
>
> _______________________________________________
> humanities-dev mailing list
> humanities-dev at lists.okfn.org
> http://lists.okfn.org/mailman/listinfo/humanities-dev

-- 
Jonathan Gray
http://jonathangray.org