We spend our last meeting reviewing the milestones we’ve created to manage the large ammount of issues we have pending in Github. The version 0.1 is working!
Notes from our July 10th 2012 meeting:
Reviewing bugs-features 1st milestone 0.1
  • Second time I run the app I get: “An error occured while installing factory_girl (3.0.0), and Bundler cannot continue.”
  • Listing of all the front pages images in display view. Fixed. Just commited.
  • Sometimes the scraper fails. Limitation of dates? For example: October 2011 fails…. fixed. Line 0 of lib/scraper.rb remove the 0, it was causing problems for months 10, 11 and 12.
2nd milestone. Review 0.1.1. July 11th 2012
  • Add the limit date for kiosko when creating a thread. Is it different for dif newspapers? I think so. It would be great to hae a messsage like: “x images from x newspapers have not been found.” We should build and scraper (for future updates) to detect when a newspaper got in kiosko.net)
  • Edit: thread features (see highlighted areas in the coding view once you come back from display). Which thing will not be edtitable? Dates and media could not be changed, so the scraper doesn’t run again.
3rd milestone 0.1.2  July 13th
Coding view:
  • Question: when coding large ammount of images, it is difficult to know where we are? which order the newspaper would appear? Order by date and not by newspaper? Let’s try to order by date.
Display view:
  • Newspaper by row https://github.com/numeroteca/pageonex/issues/31 Newspaper name in the first column.
  • Creating thumbnails for front pages https://github.com/numeroteca/pageonex/issues/22 and resize those thumbnails, not the full size pages.
  • Add link when you click on an image, so you can re-edit it.(recode images)
  • Add dates to have a reference (for each column of images shows the date of them)
  • Quantification of highlighted areas: bar graph.
  • Colors of codes in display view.
4th Online test 0.1.3 July 15th 
  • Online test? We need an online version for beta testers. Which are our need in term of server, domain… so I can prepare. We can use http://www.heroku.com/ for first test, and then start building it in our own server.
  • Compatibility with other browsers: (bootstrap itself provide this feature ) http://twitter.github.com/bootstrap/
Beta testing!
5th milestone. Dey: To be decided.
Build the tool in our own server pageonex.com
Open Id acces or Twitter…
Scraper – creating thread:
  • Open/close feature
  • Select media sources (frontpages) from different moths. Now we can only select days within a month.
  • Be able to select scraper source:
    • Kiosko.net
    • [the other built in our scrape.rb El País, NYT
  • While scraping: show files that are being downloaded/failing
  • Show which threads are opened (all threads) and be able to search.
Display view
  • Export graph and data
  • Select / unselect newspapers
  • Select order in which newspaper appear.
  • Question: how non-coder will be view the display? any diference in the links to coded image?
Ahmd should start posting more regulary:
Start with a post about the Coding view.
  • Why jquery carrousel vs. single view.
  • How highlighted ares are handled, gem used? storing coordinates? storing width-height?

And keep going with other issues to explain different decisions in the development process.

After the user creates a thread, he selects starting date and end date which could span on more than one month. The problem is that the scraping script works on one month at a time, because I’ve found a difficulty to write a method that can take the start and end date in different months and calculated the number of days between them. Because of that the days at each month changes from one year to another, that doesn’t mean it’s impossible to do, but it will add some complexity which can be avoid by, asking the user for starting day and end day for each month individuality, and run the script for each individual month.

So any suggestions on how to make this part more easy and conveniente?

issues_dates is the method which calculate the dates and return an array of the dates in this format “YYYY/MM/DD”

Look at the code of this script at https://gist.github.com/2925910