Hi Neven,

Great. How many images do you have. If you have a complete collection  
you could also use imgcollect (with some caveats though) which sort of  
supports the CTS/CITE system but not fully. Bridget could probably  
elaborate on this.

Re: your D3 examples: Sort of. In your example you deal with html  
tables (which is a bit more straight forward to select/unselect in  
D3). What you probably want is using D3 to append the image of the  
manuscript in svg and then manipulate it to your heart's content. I am  
a bit out of my comfort zone though and would have to do some trial  
and error myself. Theoretically it should work. How does EVT resolve  
this?

I am looking forward to reading the blog. Please do keep me up to  
date. I am very very interested in your progress. It seems we'd like  
to see similarly things.

Cheers,

Thomas


Quoting Neven Jovanović <[log in to unmask]>:

> Bridget and Thomas,
>
> thanks to your replies, I have been able to get the IMGSPECT from the
> Github repo going -- this seems currently the easiest way to accomplish my
> a) task. I am afraid that the SemToNotes is beyond our capabilities.
>
> I guess the way to go with d3 would be along these lines:
> <http://prcweb.co.uk/lab/selection/>
>
> The extreme importance of making the images citable (and of being able to
> forge CITE-compliant URLs for them) is very, very clear.
>
> Thanks also to all others who replied. I will summarize what we have
> learned in a blog entry and report where it can be found.
>
> Best,
>
> Neven
>
>
>
>> Neven,
>>
>> TextGrid uses SemToNotes https://github.com/HKIKoeln/SemToNotes for
>> this in combination with Sade
>> http://www.bbaw.de/telota/software/sade/documentation/documentation.
>>
>> I think you might get a, b and c from the combination of those 2
>> (maybe with or without TextGrid) but not d although as Sade is eXist
>> based it might be very easy to do d with it.
>>
>> Perseids supports a and b with SoSOL and IMGSPECT (and with CTS/CITE
>> support) and I have various bits of code for c but am also planning to
>> look at  SemToNotes soon because it supports polygons.
>>
>> Bridget
>>
>>> On Apr 28, 2015, at 6:01 PM, Thomas Koentges
>>> <[log in to unmask]> wrote:
>>>
>>> Hi Neven,
>>>
>>> As far as I know the tool you describe does not exist yet. You could get
>>> there with a combination of tools and some coding.
>>>
>>> For a) a good starting point could be
>>> http://www.homermultitext.org/hmt-doc/guides/ict.html or
>>> https://github.com/PerseusDL/imgspect and using the cite/cts
>>> architecture to link transcription and coordinates
>>>
>>> b) is not complicated once the reference link is in place
>>>
>>> For c) e.g. you could use d3 for this using the coordinates from the
>>> text/image link to draw a box when you hover over the text-passage in
>>> question with your mouse (you might want to experiment which colour and
>>> opacity works best for your needs)
>>>
>>> For d) the link pointed out in a) should help you again in combination
>>> with your programming language of choice (python or ruby jumps to mind
>>> quickly, but depending on the format of your transcription files also
>>> Xquery)
>>>
>>> The essential part is really the image citation and then you can work
>>> from there. HMT have come up with an excellent system for this. That
>>> said, once you get into collection management and there is more than
>>> only one manuscript, it becomes even more important to be very very
>>> precise with the citation system in place.
>>>
>>> Cheers,
>>>
>>> Thomas
>>>
>>> PS: I understand that you wanted to have an out-of-the-box tool, but I
>>> would be (happily) surprised if this already exists.
>>>
>>> Quoting Neven Jovanović <[log in to unmask]>:
>>>
>>>> Dear List,
>>>>
>>>> a student and I are looking for a tool which could:
>>>>
>>>> a) record a link from a line of a MS image (as a set of coordinates) to
>>>> a
>>>> line of transcription
>>>>
>>>> b) display as a web page an image map of MS page and its transcription
>>>> side by side
>>>>
>>>> c) highlight, in parallel, a line in the image and its transcription
>>>> (e.
>>>> g. on hover over image or transcription)
>>>>
>>>> d) search transcriptions for a string and display all lines where the
>>>> string appears (with links to respective images)
>>>>
>>>> In spite of the DH popularity explosion, we are having a hard time
>>>> finding
>>>> a tool that does not only a) and b), but also c) and d).
>>>>
>>>> We've looked at Image Markup Tool, TILT, Scripto, From the page,
>>>> Zooniverse / Scribe, monasterium.net, Transcribe Bentham, T-PEN -- more
>>>> or
>>>> less everything mentioned in a discussion here:
>>>> <http://digitalhumanities.org/answers/topic/collaborative-software-for-transcribing-digital-images-of-handwritten-documents>.
>>>> Nothing seems to fit our needs, however. (I've also found TextGrid, but
>>>> at
>>>> the moment it is hard for me to say whether it does what we need.)
>>>>
>>>> It appears that recently a lot of efforts were directed at
>>>> crowdsourcing
>>>> transcriptions, but not that much at displaying them.
>>>>
>>>> Any help is appreciated.
>>>>
>>>> Best,
>>>>
>>>> Neven
>>>>
>>>>
>>>>
>>>>
>>>> Neven Jovanovic
>>>> Department of Classical Philology
>>>> Faculty of Humanities and Social Sciences
>>>> University of Zagreb
>>>> Hrvatska / Croatia
>>