search

rootsdev / roots-search

Search popular genealogy websites
MIT License
24 stars 7 forks source link

Why we don't need to setup a website #41

Closed justincy closed 10 years ago

justincy commented 10 years ago

I often think about setting up a website for roots search, and many others have hinted at it. I almost started doing it today until I realized again that we don't need it. So here I'm going to document the reasons why it's tempting and then why I believe it's not necessary.

Reasons for a site

  1. There are some features that are difficult to fit into the current popup design.
    • Adding more sites. The current size of the popup limits us to 7 active sites. I want to add an options page where sites can be enabled/disabled, but I would still have to limit it to 7.
    • The current size of the text fields is too small.
    • It would be nice to turn the input fields into comboboxes (fields that you can type in or select values from a dropdown). This would allow us to support alternate names, multiple spouses, spouses maiden names, and other things like that. But we would need more space A website would give us more real estate to overcome the limitations of the popup.
  2. A site would also allow users to login to the site and track their search history
  3. We eventually want to create a plugin for Firefox. The more functionality we add to the extension the harder it will be.
  4. A site would allow us to POST data to the search page from anywhere.

Why we don't need a website

  1. We can gain additional real estate by opening a new tab instead of using a popup. The tab would be an html page bundled with the chrome extension that is basically everything the site would be.
  2. We can track history in local storage, though it wouldn't be accessible from other devices.
  3. The most complicated functionality would not be replaced by the website but remain in the extension (scraping the site and showing the page action icon in the URL bar).
  4. That's the sort of functionality that a site setup for gen-search should provide. With gen-search as a separate entity now, roots-search secret sauce is scraping information off of the websites.