An old event report doesn’t need a new font badly enough to lose its address. As I worked on the Miami Valley Orienteering Club website, the results archive was part of what the new site needed to keep: documents accumulated over years, and the links people could use to reach them.
Moving a site to WordPress doesn’t require turning every historical document into a WordPress page. For this site, a club event result had several parts that could be handled separately: the report itself, the record describing the event and linking to that report, and the page that displays the record. The report could stay where it is while the other two become easier to maintain.
For example, an event record might contain a date, a venue, and links to a recap and a results PDF. An administrator can edit that record without editing the PDF or the PHP that builds the archive page. The old document can keep its URL even when the surrounding website changes. Its typography or other styling might be a reminder of when it was written, but it will remain functional.
In this new site design, the Results record type belongs to a plugin; the theme controls its presentation. That will matter when someone eventually wants to replace the theme again. The ability to administer the club’s results shouldn’t depend on keeping a particular page design. The people who wrote those reports contributed to the club’s history, not to the useful life of a WordPress theme.
The transition also needs to work while some information is in the new system and some is still in the old one. The Results page includes a fallback list for older entries, and the importer works in batches, remembers its progress, and skips records it has already imported. Those mechanisms support a gradual transition.
Keeping the files is only part of keeping the archive. During one production check, I found that removing the helpers that generated the year-directory listings left the individual documents reachable at their URLs, but a request for a year directory was now failing. Restoring the helpers restored that route through the archive. Someone with a direct link and someone browsing by year had different results from the same collection of intact files.
That is why I keep separate checks for the documents and the ways people reach them. Comparing file hashes checks whether the preserved files match the reference copies, and following links and opening directory pages checks access. Neither substitutes for the other, and a working home page answers precious little about an old results archive.
Even the file count needed an investigation. An earlier inventory had 822 files; the reviewed static archive contained 819. The difference turned out to be three old transfer logs deliberately excluded after review, not three missing event reports. The site also needed 28 directory helpers, counted separately from the historical documents. Before treating a smaller count as data loss (or a larger one as success), you have to find out what is being counted.
Preserving a document doesn’t fix everything about it. An old PDF may still be difficult to read on a mobile device or use with assistive technology. Keeping it available preserves the opportunity to do that work.
Before redesigning an organization’s website, identify the material people still need, the links they use, and who will maintain it. Then decide what needs conversion and what can stay in its existing form. A new design shouldn’t disconnect users from their history.
—jhunterj