{"id":5730,"date":"2026-11-21T01:50:00","date_gmt":"2026-11-21T01:50:00","guid":{"rendered":"https:\/\/www.devopssupport.in\/blog\/?p=5730"},"modified":"2026-11-21T01:50:00","modified_gmt":"2026-11-21T01:50:00","slug":"where-to-keep-assumptions-compared","status":"publish","type":"post","link":"https:\/\/www.devopssupport.in\/blog\/where-to-keep-assumptions-compared\/","title":{"rendered":"Where Should Written Assumptions Live? Four Homes Compared"},"content":{"rendered":"<p>Teams that run support and managed operations quickly discover that the hard part of writing down assumptions is not the writing. It is deciding where the list should live so that it gets read, updated and trusted. This comparison looks at four common homes for an assumptions list, judges each on the same criteria and suggests which fits which kind of work. The goal is a choice you can defend, not a universal winner.<\/p>\n<h2>The criteria that matter<\/h2>\n<p>Before comparing, fix the yardsticks. Four are worth using for any home of a list of beliefs:<\/p>\n<ol>\n<li><strong>Visibility:<\/strong> will the people who depend on the beliefs actually see them?<\/li>\n<li><strong>Freshness:<\/strong> how easy is it to update the line when something changes?<\/li>\n<li><strong>History:<\/strong> can you see what the list said last month and why it changed?<\/li>\n<li><strong>Effort:<\/strong> how much friction does adding a line add to the day?<\/li>\n<\/ol>\n<p>A home that scores well on three and badly on the fourth tends to be abandoned within weeks, so effort deserves honest weight.<\/p>\n<h2>Option one: the chat thread<\/h2>\n<p>Many assumptions are first stated in a team chat: &#8220;I&#8217;m assuming the failover target has the same capacity.&#8221; It is fast, it needs no ceremony and it captures the belief at the moment it forms. That is its strength.<\/p>\n<p>Its weaknesses are serious. Visibility decays within hours as messages scroll away. Freshness is poor, since nobody edits a chat message to reflect a later discovery. History exists only as a search problem. Chat is useful as a seed, a place where a belief is first spoken, but a poor archive. A good habit is to copy the line into a durable home as soon as it is spoken aloud.<\/p>\n<h2>Option two: a field in the work ticket<\/h2>\n<p>Adding an &#8220;Assumptions&#8221; section to each ticket places the beliefs next to the task, in front of the person doing the work and the person reviewing it. Visibility is high for the life of that ticket. Effort is low, because the person is already there.<\/p>\n<p>The drawbacks appear over time. Tickets close, and with them the practical visibility of their contents. A belief that spans several tickets, such as the capacity of a shared service, gets repeated in each and drifts into inconsistent versions. History is acceptable, since ticket systems keep edit trails, but it is scattered. This option suits short pieces of work and incident follow-ups well.<\/p>\n<h2>Option three: a table in the design or service document<\/h2>\n<p>A dedicated table in a document that describes how a service is built and run gives the list a stable address. It naturally gathers the long-lived assumptions about a service: expected load, dependencies, who approves changes, which maintenance windows apply.<\/p>\n<p>On visibility, it depends on whether people open the document. On freshness, it is only as good as the team&#8217;s editing discipline. On history, most document tools offer version lists, which is a respectable record. The cost is moderate: adding a row means leaving the ticket and finding the page. This option suits ongoing managed services, where the same beliefs apply to many tasks over a long time.<\/p>\n<h2>Option four: a standalone log kept in the repository<\/h2>\n<p>A plain file in the same repository as the code or configuration, changed through the ordinary review process, offers the strongest history, since every edit has an author, a date and a reviewer. Updates travel with the change they relate to, so freshness can be excellent when the team treats the file as part of the work.<\/p>\n<p>Its visibility is the weak spot for non-engineers, who rarely browse repositories. Effort is low for developers and high for anyone else. This option works best where the audience is mostly engineers and where the beliefs relate directly to code or configuration.<\/p>\n<h2>Side by side<\/h2>\n<ul>\n<li><strong>Chat thread:<\/strong> visibility short-lived, freshness poor, history weak, effort lowest.<\/li>\n<li><strong>Ticket section:<\/strong> visibility high but local, freshness fair, history scattered, effort low.<\/li>\n<li><strong>Service document table:<\/strong> visibility medium, freshness depends on discipline, history decent, effort moderate.<\/li>\n<li><strong>Repository file:<\/strong> visibility narrow, freshness strong, history strongest, effort low for engineers only.<\/li>\n<\/ul>\n<p>No option dominates. Which one wins depends on the nature of your work. A team that mostly handles quick incidents will lean on tickets, while one that operates a handful of long-lived services for clients will prefer the document table, perhaps backed by the repository file for configuration-level beliefs.<\/p>\n<h2>A combination that tends to work<\/h2>\n<p>Many support teams settle on a two-layer arrangement. Long-lived beliefs about a service sit in its document table. Beliefs specific to a piece of work sit in the ticket, with a link back to the table whenever they depend on it. Chat is allowed as a notepad, with one rule: anything that survives a day moves to a durable home. This design keeps effort modest and avoids duplicating the same belief in a dozen places.<\/p>\n<p>The broader principles that shape this kind of arrangement, such as keeping feedback short and work visible, are discussed in the piece on <a href=\"https:\/\/www.devopssupport.in\/blog\/devops-principles-for-effective-software-delivery\/\">principles that make software delivery effective<\/a>. For engineers who are moving from hands-on operations toward code-driven practice, the article about <a href=\"https:\/\/www.devopssupport.in\/blog\/infrastructure-as-code-to-sre-why-operations-engineers-should-embrace-devops\/\">why operations people benefit from adopting DevOps habits<\/a> offers helpful context for the repository option.<\/p>\n<h2>A quick scenario to test the choice<\/h2>\n<p>Suppose a support team looks after a customer&#8217;s reporting database and an overnight export job. Three beliefs matter: the export finishes before the customer&#8217;s morning meeting, the database replica lags by only a short interval, and the customer&#8217;s contact approves emergency changes by phone. The first two are long-lived and technical, so they belong in the service document, where every engineer on rotation will meet them. The third is about a person and tends to change when staff change; it belongs in the same table, with a recheck date, rather than in a ticket nobody will reopen.<\/p>\n<p>Now suppose a one-off migration ticket arrives for the same customer. Its specific beliefs, such as the size of the data to move and the length of the agreed window, belong in that ticket. When the ticket closes, the team scans its list and promotes anything of lasting value up to the service document. That promotion step is cheap and prevents lessons from being buried.<\/p>\n<h2>Outside reading<\/h2>\n<p>Several training sites cover documentation habits for operations teams. NoOpsSchool, based in India, describes on its own pages <a href=\"https:\/\/noopsschool.com\/\">certification courses with accompanying consulting and services for IT professionals<\/a>, and may serve as one more reference point when you look at how technical teams organise what they write.<\/p>\n<h2>Verdict<\/h2>\n<p>Choose the home that your readers already visit every day. If your engineers live in tickets, begin there. If your clients read a service document, begin there. Pick one for the next month, add five lines, and note after thirty days which of the four criteria let you down. That review will tell you more than any comparison, including this one.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Chat, tickets, service documents and repository files all can hold a list of assumptions. We compare them on visibility, freshness, history and effort.<\/p>\n","protected":false},"author":8,"featured_media":5731,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[3948],"tags":[],"class_list":["post-5730","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-devops-support"],"_links":{"self":[{"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/posts\/5730","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/comments?post=5730"}],"version-history":[{"count":1,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/posts\/5730\/revisions"}],"predecessor-version":[{"id":5826,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/posts\/5730\/revisions\/5826"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/media\/5731"}],"wp:attachment":[{"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/media?parent=5730"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/categories?post=5730"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.devopssupport.in\/blog\/wp-json\/wp\/v2\/tags?post=5730"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}