Team:

Thanks for getting together to discuss how the Job Bank process can be improved. We had a lot of good ideas, and it was good to hear everyone’s visions of what the Job Bank can be in the future.

The number-one goal at hand, as stated by Steve, is to reduce the amount of manual labor involved in the weekly production of the Job Bank email. It would be best if much of the process can be automated with tools.

If we can reach consensus on this, it will be much easier to move forward.

Assuming we all agree, then it stands to reason that reproducing the weekly email, in its current format, would be acceptable. (The fewer changes, the less upheaval.)

(I think members are by and large satisfied that there actually is a newsletter. Suggestions for improving the format should be secondary.)

With this in mind, I propose this as the first phase of the project:

  1. The online submission form is modified to put the job details into the database.
  2. Add validation to help reduce bogus or incomplete submissions.
  3. Convert text to ASCII.
  4. When a job is submitted, an email goes out to the Job Bank volunteers. The email has a link to a Web page that shows what got submitted. There’s a button on the page to approve the submission. The volunteer also should have the ability to edit the job description (as volunteers currently can do so).
  5. On Sunday, an automated script compiles all the jobs submitted during the week (after the publishing of the previous week’s edition) into an email and sends it to the Job Bank volunteers.
  6. A Job Bank volunteer emails the jobs as it is currently done today.

Phase 2:

  1. On Sunday, instead of emailing the compiled jobs to the volunteers, directly email the jobs to the subscribers.

Phase 3:

  1. Modify the form to add a field for a URL for the job, so that users can click directly on the link to view the “official” job posting and apply.
  2. If the field is left blank, generate a URL for a BayCHI-hosted page for the job.
  3. Create an index page on the BayCHI site for recent jobs. Perhaps group them by week, and go only 30 days back.
  4. The index page will link to a “details” page for each posting.
  5. The index page may have jobs that have been added since the last emailing. The email is a snapshot of the jobs submitted within the last seven days.
  6. The email will contain a randomized URL to the index page. The URL will be good for a limited time, say 30 days. The index page URLs included on all Job Bank emails within the last 30 days will go to the same page: the Job Bank index page hosted on baychi.org. (There will be no public links to that page.) URLs from older emails will not “forward” to the index page anymore.

Phase 4:

  1. Set up self-service for members to subscribe to and unsubscribe from the newsletter, and update their email addresses.

Phase 5:

  1. Consider reformatting the email with HTML. Even though the branding benefits for a redesign may be limited, as Steve explained, the readability can be improved by following some design principles.

    Steve Williams:
    Not readability, surely. The best readability is always achieved by letting each reader choose the font and size of text. HTML can only interfere with the reader’s preference.

    Rather, HTML might allow a more usable overall layout. Within that layout, readability almost certainly will be reduced, but that’s fine: Our goal is no longer to email full job postings that must be read. Rather, we want to send brief summaries and pointers to the full job postings on the web.

    As long as the email is skimmable, searchable, absorbable, it need not be readable.

  2. HTML email will enable the jobs listed at the top to be hyperlinked to the job description somewhere lower in the email, which can get very long. (If we do this, there should be a “back to top” link beneath each job description.)

    Steve Williams:
    No, we specifically do NOT want to include the job descriptions in the email. As of Phase 3, the email carries only metadata. Each job summary in the email links to the job posting on the web.

    Even if we go to HTML, it’s important to keep the Job Bank email as few bytes as possible.

I propose we just do phase 1 for now, keeping in mind the primary goal of reducing volunteer hours. I propose we do not try to bite off phase 2 and beyond right away, or wait till we come up with the perfect plan, because that will delay phase 1. We should keep the scope small for now.

What does everyone think?

Thanks,
Andrew.