Posts tagged release-notes

October 2017 Release Notes

Highlights

  • To accomplish our FY 2018 goals, we continued backend development that will allow your agency content to be sent directly into our indexes. In October, our team completed work that enables our system to process PDFs, Word documents, and Excel files.

Chores

  • We completed our transition to Chef 12.
  • We updated our program website to be https://search.gov/. As part of that process, we changed our website favicon to our new logo, updated references to our program on individual pages, and created a new domain mask.
  • We updated the attribution at the bottom of search results pages that display content from our indexes. Those pages now say “Powered by Search.gov.”
  • We began transitioning our repos to use Circle CI.

Fixes

  • Our hosted search results pages are 508 compliant, and we undergo regular review to ensure compliance. This month, we added underlining to the links that are displayed in the headers, footers, and navigation bars on our search results pages. The underlining appears when you hover over the link, and indicates that the links are not regular text.
    Screenshot of USA.gov search results page with green box around facets

Page last reviewed or updated:

September 2017 Release Notes

Highlights

  • We announced the changes coming to our program in FY 2018. Read our blog post and FAQ for more information.

  • To accomplish our FY 2018 goals, we continued backend development that will allow your agency content to be sent directly into our indexes. In September, our team investigated multiple tools that can fetch website content, parse it, and send it into our Elasticsearch indexes. We also completed work that will ensure our new system follows “noindex” meta tags and headers if they are present on your website.

Chores

  • We upgraded our Ruby version to 2.3.5.

  • We prepared to transition our program website to the search.gov domain.

  • We deactivated any users who had not accessed the Search Admin Center within the past year. We notified inactive users prior to deactivating their accounts.


Page last reviewed or updated:

August 2017 Release Notes

Highlights

  • Agencies can send content directly to our indexes via our i14y API. Previously, each document had to have a created date. Now, this field is optional. Review our technical documentation for more information.

  • We continued work on backend features that will provide new ways for customer content to be sent directly into our own indexes. In August, our team created a rake task that will allow our system to fetch, parse, and index a list of URLs. We also investigated the spellchecker functionality in Elasticsearch. This is a long term project that extends through the rest of this calendar year and into 2018. Stay tuned!

Chores

  • We completed our upgrade to Ruby 2.3.

  • We continued work that will upgrade our server recipes to Chef 12.


Page last reviewed or updated:

July 2017 Release Notes

Highlights

  • We’ve begun work on backend features that will provide new ways for customer content to be sent directly into our own indexes. In July, a new index was set up in our system. We also investigated ways to fetch, parse, and send data to our indexes. This is a long term project that extends through the rest of this calendar year and into 2018. Stay tuned!

Chores

  • We wrapped up refinements to our bot-blocking feature that was released in May.

  • We continued work that will upgrade our server recipes to Chef 12 and our Ruby version to 2.3.

Fixes

  • As part of our compliance with Section 508 law, we increased the size of facet text on our search results pages. The guideline is that text should be at least 13 pixels, and our facet text is now 13.125 pixels.

Screenshot of USA.gov search results page with red box around facets


Page last reviewed or updated:

June 2017 Release Notes

Highlights

  • Take complete control of your search results! We released the dev Drupal 8 version of our Drupal module. The module hooks your CMS into i14y, our content indexing API. Through i14y, you can send content directly to our indexes. No commercial web results are served: instead, searchers will see content exactly as you’ve sent it to us. Go to our Drupal project page to try it out, and reach out to our team with questions.

  • Our developer team has begun work on backend features that will provide new ways for customer content to be sent directly into our own indexes. This will be a long term project that extends through the rest of this calendar year and into 2018. Stay tuned!

Chores

  • We continue to refine our bot-blocking feature that was released in May. We are excited at the success of the feature and are continuing to stop new ‘junk’ queries from being entered into your search forms.

  • We began updating our server recipes to Chef 12 and our Ruby version to 2.3.

Fixes

  • With our new bot blocker feature, we discovered that the CAPTCHA page didn’t work for human searchers who don’t have Javascript enabled, and searchers did not see an error. We added an error message to instruct these searchers to temporarily enable Javascript in order to complete the CAPTCHA prompt.

Page last reviewed or updated:

May 2017 Release Notes

Highlights

  • What type of traffic do we not like in our system? Bots. Bots run junk queries on your search boxes, gobbling up our query purchases with the commercial search engines. To battle the bots, we launched a bot-blocking feature:
    • If an IP address sends high levels of traffic to a particular search site in a short amount of time, the searcher will be asked to confirm that they are human via CAPTCHA (screenshot below).
    • The CAPTCHA is unlikely to trigger for humans.

Screenshot of bot blocking message and CAPTCHA

Chores

  • Our developer team continues to investigate and roadmap the best way to get customer content into our own indexes. This will be a long term project that extends through the rest of this calendar year and into 2018. There are several advantages to using our own indexes:
    • Customers can have complete control over what content is included in their search results.
    • Customers don’t need to rely on results from commercial indexes, which can be out of date or have unpredictable titles/descriptions.

Page last reviewed or updated:

April 2017 Release Notes

Highlights

  • We deactivate user accounts for customers who are no longer at their agency, do not have a search site set up, have never verified their email, or have not logged in to their account for more than a year. Previously, deactivated users who attempted to log-in to our system and/or reset their passwords were redirected to the USA.gov homepage. Now, deactivated users see a message, alerting them to contact our team.
  • We hosted a webinar covering the current ways that customers can directly index their content with our service: via Supplemental RSS feed, i14y API or via the Drupal module that connects with our API. We’ll post the recording in the Training section of our website as soon as it is available.

Chores

  • Our developer team began a sprint investigating the best way to get customer content into our own indexes. There are several advantages to using our own indexes:
    • Customers can have complete control over what content is included in their search results.
    • Customers don’t need to rely on results from commercial indexes, which can be out of date or have unpredictable titles/descriptions.

Fixes

  • Our Jobs API was briefly offline on Thursday, April 27th (from approximately 4:50 pm - 5:40 pm). A DNS update made by another program in our division caused the outage, which was immediately caught by our alerting system and rectified.

Page last reviewed or updated:

March 2017 Release Notes

Highlighted Features

  • Customers can use our Supplemental RSS Feed and Supplemental URLs feature to add missing content to their search results or to have complete control over what content appears. Previously, every item had to have a description. Now, descriptions are optional. If there is no description for an item, we’ll display snippets taken from an item’s full text. If a description is provided, we’ll show the description if it has a match with the query; otherwise, we’ll show snippets from the full text. Note: descriptions are still required for our main RSS feed feature.

Chores

  • As of March 1st, we only support federal and District of Columbia websites. We wrapped up our behind-the-scenes work that facilitated this transition.
  • Google announced that it is deprecating the Google Site Search API. We use the GSS API to serve Google results in the limited cases where Bing results are not sufficient. We’ve renewed our license and will continue to use GSS until its end date of March 31, 2018. We are investigating how to best mitigate the effect of this deprecation.
  • A new help manual page has been added for our Analytics Alert feature. We also revised the pages for adding supplemental content to your search site and filtering URLs.

Fixes

  • Users who were deactivated from our system still received Daily Snapshot emails of their site’s activity. This has been corrected, and deactivated users no longer receive those emails.
  • On March 17th, our system was offline for 1 hour. We traced the problem to a specific configuration in our new logging mechanism. We mitigated the issue and will ensure the long term solution is stable.

Page last reviewed or updated:

February 2017 Release Notes

Highlighted Features

  • If you change your email address, we will now send a verification link to the new address.
  • We now offer a Kazakh language search results page.

Chores

  • As of March 1st, we only support federal and District of Columbia websites. We’ve done a lot of behind-the-scenes work to support our non-federal customers during the past few months of transition.
  • We’ve added our active repos to Code.gov, a website promoting good practices in code development, collaboration, and reuse across the federal government.
  • Our sysadmins reduced the size of our Elasticsearch server cluster, saving money while maintaining system integrity.
  • We removed old code from our system, wrapping up tasks from our December transition to Bing’s newest API.

Fixes

  • We fixed internal reports that allow us to track when agencies launch our service on their sites.
  • We fixed a bug that affected the indexing of some Twitter accounts.

Page last reviewed or updated:

January 2017 Release Notes

Highlighted Features

i14y Drawer Management

We updated the process for removing i14y Drawers from a search configuration.

  • If you remove a drawer that is associated with only one search configuration, it will be deleted from our system.
  • If you remove a drawer that is attached to multiple search configurations, it will be removed from only the one configuration and will remain attached to the others.

New Results Page Platform Launches

We launched our new JavaScript search results page, which is serving the search results on IRS.gov.

New JavaScript results page

Chores

  • Have your RSS feeds transitioned from HTTP to HTTPS? Our system now recognizes that transition automatically.
  • We indexed and support the search for whitehouse.gov. We prepared indexes to go live at noon January 20, for the obamawhitehouse.archives.gov sites, as well as the new content on whitehouse.gov.
  • We’ve updated the process around password reset links.
    • Password reset links are only valid for one hour: previously, if you requested a password reset link more than once within the hour, the first link would expire. This caused confusion for many users who accidentally hit the reset link twice and were encountering expired links.
    • Now, we will resend you the first, valid link for a period of one hour, regardless of how many times you hit the reset button.
    • We’ve also updated and clarified the message you see if you click an expired link.
  • We finalized adding all our active repos to the Code.gov inventory.

Fixes

  • Our excellent asset caching was inadvertently prolonging transient errors by serving up cached error responses. We’ve fixed this so errors will really go away when they’ve cleared.
  • We also added a response code passthrough to make sure that, in our multiserver environment, we don’t end up returning 200 responses just because an error page is working just fine. We are now passing through the response code reported by the error page.
  • We fixed a bug where some password reset tokens were delivering users to the login page, when they should have been delivered to the password reset page.
  • We fixed an error that was interfering with our import of Federal Register documents for indexing.

Page last reviewed or updated: