otsukare Thoughts after a day of work

Steps Before Considering a Bug "Ready for Outreach"

Sometimes another team of Mozilla will ask help from Webcompat team for contacting site owners to fix an issue on their Web site which hinders the user experience on Firefox. Let's go through some tips to maximize the chances of getting results when we outreach.

Bug detection

Bug

A bug has been reported by a user or a colleague. They probably had the issue at the moment they tested. The source of the issue is still quite unknown. Network glitch, specific addon configuration, particular version of Firefox, broken build of Nightly. Assess if the bug is reproducible in the most neutral possible environment. And if it's not already done, write in the comments "Steps to reproduce" and describe all the steps required to reproduce the bug.

Analyzing the issue

Bug

You have been able to reproduce. It is time to understand it. Explain it in very clear terms. Think about the person on the other hand who will need to fix the bug. This person might not be an English native speaker. This person might not be as knowledgeable as you for Web technologies. Provide links and samples to the code with the issue at stake. This will help the person to find the appropriate place in the code.

Providing a fix for the issue

Bug

When explaining the issue, you might have also find out how to fix it or at least one way to fix it. It might not be the way the contacted person will fix it. We do not know their tools, but it will help them to create an equivalent fix that fits in their process. If your proposal is a better practice explained why it is beneficial for performance, longevity, resilience, etc.

Partly a Firefox bug

Bug

The site is not working but it's not entirely their fault. Firefox changed behavior. The browser became more compliant. The feature had a bug which is in the process of being fixed. Think twice before asking for outreach. Sometimes it's just better to push a bit more on fixing the bug in Firefox. It has more chances to be useful for all the unknown sites using the feature. If the site is a big popular site, you might want to ask for outreach, but you need a very good incentive such as improving performances.

Provide a contact hint

Bug

If by chance, you already have contacts in this company, share the data, even try directly to contact that person. If you have information that even bookies don't know about the company, be sure to pass it on for maximizing the chances of outreach. The hardest part is often to find the right person who can help you fix the issue.

Outreach might fail

And here the dirty secret: The outreach might now work or might not be effective right away. Be patient. Be perseverant.

Bug

Fixing a Web site costs a lot more than you can imagine. Time and frustration are part of the equation. Outreach is not a magical bullet. Sometimes it takes months to years to fix an issue. Some reasons why the outreach might fail:

  • Impossible to find the right contacts. Sometimes you can send bug reports through the official channels of communications from the company and have your bug being ignored, misunderstood, unusual. For one site, I had reported for months through the bug reporting system until I finally decided to try a back door with emailing specifically a developer I happened to find the information online. The bug was fixed in a couple of days.
  • Developers have bosses. They first need to comply with what their bosses told them to do. They might not be in a very good position in the company, have conflicts with the management, etc. Or they just don't have the freedom to take the decision that will fix the issue, even a super simple one.
  • Another type of bosses is the client. The client had been sold a Web site with a certain budget. Maintenance is always a contentious issue. The Web agencies are not working for free and even if the bug is there in the first place. The client might not have asked for them to test in that specific browser. Channeling up a bug to the client will mean for the Web agency to bill the client. The client might not want to pay.
  • Sometimes, you will think that you got an easy win. The bug has been solved right away. What you do not know is that the developer in charge had just a put a hack in his code with a beautiful TOFIX that will be crushed at the next change of tools or updates.
  • You just need to upgrade to version X of your JS library: Updating the library will break everything else or will require to test all the zillion of other features that are using this lib in the rest of the site. In a cost/benefit scenario, you have to demonstrate to the dev that the fix is worth his time and the test.
  • Wrong department. Sometimes you get the press service, sometimes the communications department, sometimes the IT department in charge of the office backend or commercial operations systems, but not the Web site.
  • The twitter person is not techy. This happens very often. With the blossoming of social managers (do we still say that?), the people on the front-line are usually helpless when it's really technical. Your only chance is to convince them to communicate with the tech team. But the tech team is despising them because too often they brought bugs which were just useless. If the site is an airline company, a bank, a very consumers oriented service, just forget trying to contact them through twitter.
  • The twitter person is a bot. Check the replies on this twitter account, if there is no meaningful interaction with the public, just find another way.
  • You contacted. Nothing happened. People on the other side forget. I'm pretty sure you are also late replying this email or patching this annoying bug. People ;)
  • The site is just not maintained anymore. No budget. No team. No nobody for moving forward the issue.
  • You might have pissed off someone when contacting. You will never know why. Maybe it was not the right day, maybe it was something in your signature, maybe it was the way you addressed them. Be polite, have empathy.

In the end my message is look for the bare necessities of life.

Bugs images from American entomology : or description of the insects of North America, illustrated by coloured figures from original drawings executed from nature. Thanks to the New-York Public Library.

Otsukare!