<div dir="ltr">Hi Sasha,<div style>This is not yet on the issue tracker but is definitively an important step forward that it has came out several times (like when I presented in <a href="http://mrap.info/index.html">http://mrap.info</a>). Researchers need a tool that not only helps them code, but that makes part of the pre-coding for them.</div>
<div style><br></div><div style>Do you want to add it to the issue tracker? I can also do it. We'll discuss how difficult to develop this is soon.</div><div style><br></div><div style>thanks,</div><div style>p</div></div>
<div class="gmail_extra"><br><br><div class="gmail_quote">On Fri, Apr 5, 2013 at 2:19 PM, Sasha Costanza-Chock <span dir="ltr"><<a href="mailto:schock@mit.edu" target="_blank">schock@mit.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi Pablo, this is great! I tried out the dev version last night with (mostly) success :)<br>
<br>
I have a quick question re: user story I think is pretty crucial to solve:<br>
<br>
I'm a user, and I want to search for terms on newspaper front pages, so I know what date range to limit my pageonex project to.<br>
<br>
To solve this, ideally we would have to integrate with a search tool that knows whether stories are on the front page or not. Lexis Nexis can do this, for example, with NYT (limit the search to page A1), but I don't know whether it also does so for all papers. If we got integration of a search tool that does this, actually, we could save project creators and coders a huge amount of work.<br>
<br>
If that's not possible, we might at least provide people a link (or an iframe or modal box?) that allowed them to search on page1 using an external search tool (the hacky solution to the user story).<br>
<br>
Anyway, just wondering whether this is in the issue tracker yet. Thanks!<br>
sc<div class="im"><br>
<br>
<br>
<br>
<br>
On 04/05/2013 10:08 AM, pablo rey wrote:<br>
</div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div class="im">
Hi everyone,<br>
After last commits we are about to launch the beta site at <a href="http://pageonex.com" target="_blank">pageonex.com</a><br>
Congrats everyone for the hard work!<br>
<br>
What do we have so far:<br>
<br>
Installed developers version at <a href="http://dev.pageonex.com" target="_blank">dev.pageonex.com</a> [don't share it publicly<br>
yet, although this list is public :-/ ] that includes all the new features<br>
[add disclaimer saying that it is a dev version, data might be erased].<br>
Among many other things<br>
<br></div>
- multiple area selection<br>
- export to ods and json [fix bug]<br>
- add/delete topics<br>
- clean database structure and code<br>
- new home + thread index view<br>
<br>
We are working on:<br>
<br>
- Fix last bugs and tag this as version beta 1.0?<br>
- Test with real cases the dev site<br>
- Once ready launch the beta version publicly at <a href="http://pageonex.com" target="_blank">pageonex.com</a><br>
- Pre-load this version with some key threads (from the dev version and<div class="im"><br>
previous heroku like the thread about<br></div>
corruption<<a href="http://pageonextester4.herokuapp.com/users/numeroteca/threads/noticias_sobre_corrupcion_en_spain" target="_blank">http://<u></u>pageonextester4.herokuapp.com/<u></u>users/numeroteca/threads/<u></u>noticias_sobre_corrupcion_en_<u></u>spain</a>><div class="im">
<br>
)<br>
<br>
Next steps:<br>
<br>
I've listed here some key points, to have a general view of what we can do<br>
in the following moths. We could discuss them on this Monday developers<br>
session.<br>
<br></div>
*Scraping*<div class="im"><br>
We are currently scraping from <a href="http://kiosko.net" target="_blank">kiosko.net</a>, future steps would include other<br>
data sources such as:<br>
<br></div>
- newseum (<a href="http://www.newseum.org/todaysfrontpages/" target="_blank">http://www.newseum.org/<u></u>todaysfrontpages/</a>) I am contacting<div class="im"><br>
them to see what is possible to get from them<br></div>
- local files: images uploaded by users<br>
- newspapers: set up scrapers for the different newspapers (gdoc that<div class="im"><br>
list different newspaper front page data to help bulid a<br></div>
script<<a href="https://docs.google.com/spreadsheet/ccc?key=0AupjZBpCwY8UdEgwUndSeHp5bjBMRHlJME1TSkZRZkE#gid=0" target="_blank">https://docs.google.<u></u>com/spreadsheet/ccc?key=<u></u>0AupjZBpCwY8UdEgwUndSeHp5bjBMR<u></u>HlJME1TSkZRZkE#gid=0</a>><div class="im">
<br>
)<br>
<br>
The user would be able to select from different data sources when starting<br>
a thread.<br>
<br></div>
*Coding*<br>
Basic features:<br>
<br>
- Fix slow behavior when many images in the thread with:<br>
- ajax<br>
- resize of images to show smaller images in the display view<br>
- Allow multicoder (sharing thread ownership)<br>
<br>
Others<br>
<br>
- Allow simultaneous coding<br>
- Add annotation features (for days, for images)<br>
- Capability for magnet areas to avoid overlapping of areas<br>
- allow connection of areas to form one "area-article", this will lead<div class="im"><br>
to be able have the "count" of articles per day/newspapers in one topic.<br></div>
- Allow multiple taxonomies (Ex: one taxonomy for theme + taxonomy for<div class="im"><br>
frame analysis<br>
<a href="http://numeroteca.org/2013/02/06/3-steps-to-measure-the-corruption-coverage-in-spain/" target="_blank">http://numeroteca.org/2013/02/<u></u>06/3-steps-to-measure-the-<u></u>corruption-coverage-in-spain/</a><br>
)<br>
<br>
<br></div>
*Display*<div class="im"><br>
Test D3 to allow richer visualizations and direct svg integration:<br>
<br></div>
- bar char split by newspapers<br>
- live selection of newspapers to modify on the fly which newspapers<div class="im"><br>
show in the bar chart<br>
<br>
Enable the selection of only certain days for the display view (different<br>
form the original time frame of the thread). Very long threads<br>
have difficulties for being displayed. In the past, a year long study<br>
implied the creation of 12 threads. This new feature would enable the<br>
creation of 1 thread and then visualize only certain parts of it.<br>
<br></div>
*Platform*<br>
<br>
- Allow search and tagging of threads<br>
<br>
*Website*<div class="im"><br>
I am moving the blog to a location in the <a href="http://pageonex.com" target="_blank">pageonex.com</a> domain to have<br>
everything integrated.<br>
<br></div>
*User driven development*<div class="im"><br>
People using the beta site would find bugs and will have needs.<br>
<br></div>
- Encourage the use of the issue tracker and users mail list to report<br>
ideas, suggestions, needs...<br>
- Contact all the people that used previous versions of PageOneX in<div class="im"><br>
heroku to test the beta site.<br>
<br>
A lot of ideas, and much more that do not fit in this email. Happy to see<br>
that we'll be on the air soon.<br>
<br>
Happy spring,<br>
p<br>
<br>
</div></blockquote><span class="HOEnZb"><font color="#888888">
<br>
<br>
-- <br>
Sasha Costanza-Chock<br>
Assistant Professor of Civic Media<br>
<br>
Comparative Media Studies<br>
Massachusetts Institute of Technology<br>
<a href="tel:1.607.351.5559" value="+16073515559" target="_blank">1.607.351.5559</a> | schock AT <a href="http://mit.edu" target="_blank">mit.edu</a> | @schock | E15-322<br>
<a href="http://cms.mit.edu" target="_blank">http://cms.mit.edu</a> | <a href="http://civic.mit.edu" target="_blank">http://civic.mit.edu</a> | <a href="http://schock.cc" target="_blank">http://schock.cc</a><br>
<br>
</font></span></blockquote></div><br></div>