• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 205
  • Last Modified:

website content maintenance

is there any sort of terminology used by web developers to cover how you keep content current - i.e. remove any old irrelevant content, update with new relevant content, update and maintain links etc.

We have a number of websites developed by 3rd parties and are interested in auditing there web maintenance processes, but I am unsure of the exact wording for this activity - or whether there are any frameworks/best practices checklists web dev teams work towards to keep their sites up to date and current, that we could use as a checklist to see how well they are doing so. If such frameworks and checklists do exist - the details would help no end.
2 Solutions
Terry WoodsIT GuruCommented:
When it comes to keeping content current, I'd call it content management. It makes complete sense to call it that when you're using a content management system (CMS) to do the work, and even if you're not.

Managing servers, software, security and hosting to keep the content online is different though.

At the web agency I'm working for, I've recently standardised processes for security and software maintenance. We use an extensive go-live checklist to review sites prior-to and after go-live too. Sadly I can't share documentation here, as it's proprietary.

Processes are always going to be dependent on the multiple layers of the stack of software used to run a site, so what I've got would like not fully work for you anyway. It's easy to develop processes; start by documenting what you know needs to be in it, and then just add to it as you discover things that aren't in it already.
Steve BinkCommented:
I'm not in charge of such things at my work, but my boss (let's call him Alfred) has several processes he follows obsessively.  I'll cover two projects - one internal and one public site.  We have an almost team environment (multiple developers, but we don't generally collaborate on individual tasks), and generally use GitHub to manage our repos.

For the public site, we have the additional complication of a third-party contractor.  They maintain the canon branch, while we maintain our development environments and GH repos.  As a developer, I code locally until it works.  Then I push to a VM environment at the host.  The VM is always seeded with the live code base and database, then our patches are applied.  The VM is tested by people not me, including the projects manager and the contractor.  Once accepted, we create a PR into our master GH branch.  Once the PR is complete, the contractor schedules and manages the merge into live.  

For the internal app, Alfred has established three primary branches - prod, test, and dev.  My team and I code and test locally.  Once our work is accepted, it is pushed to dev.  We don't use PRs like we should, so we just push to the branch.  There, it is tested by people not me, including Alfred.  Once the powers that be have basic satisfaction it isn't going to end our world, it is pushed to test for department-wide testing.  This includes both manual testing and some automated data entry tests for basic functions.  Once it is determined to be production-worthy, it is pushed to prod, and actively monitored for at least the next week.  The app serves 80-something separate offices, so any faults are going to show up pretty quick, and our users are very vocal.  If anything horrible happens, prod is reverted, and the team goes back through debug-reproduction-fix-test until it is right.  Note that only prod is reverted - dev and test maintain the full history, warts and all.  For successful pushes, prod branch is tagged with version markers.  Alfred maintains the version count and change logs.

Aside from the coding, Alfred also manages our internal servers.  He does not handle networking, hardware, etc. - just pure server admin stuff.  We run LAMP, Tomcat, and several other (generally open source) applications.  Alfred is subscribed to the various Twitter, news feeds, email lists, GH repos, etc., such that when any app we use receives an update, he is alerted.  He runs through the change log to get a general idea of the impact we can expect.  Once he is satisfied a transition can be made, he'll update the software on some of the test servers and let us try to break it.  Once every one is comfortable, the same updates are applied to dev, where the developer and testing teams actively use it.  Once he's comfortable THAT is OK, prod is finally updated, and the developers usually update their locals at the same time.

Alfred also maintains and watches statistics for server health.  On a daily basis, he monitors server traffic, server load, memory usage, feature usage, general responsiveness, and the various error logs for our stack.  If anything untoward pops up, he tries to resolve it.  If it is a longer fix, or requires a software patch (e.g., memory leak), he will delegate a ticket to one of the developers.

If these processes are formally documented, I've never seen it.  AFAIK, this is just Alfred's regimen.  I have the sneaking suspicion that if Alfred were to suddenly disappear tomorrow, things would come to a screeching halt in my office.  Until then, he does an amazing job keeping us productive... by far, the best manager I've worked with.  

While my scenario is mostly focused on code, content management is going to follow some of the same paths.  Instead of a development process, it's a writing process.  Testing becomes editorial review.  Versioning is usually managed inside the CMS.  If your CMS doesn't have versioning (i.e., you're doing it wrong), the robust backup system you undoubtedly have in place serves that purpose in a much more painful and tedious way.  

Which reminds me of one more item - backups.  Take them regularly and keep them as long as possible, plus one month.  Make sure you have at least weekly full backups with daily differentials.  You should have a milestone backup prior to any significant push.  And remember: your backups are only as good as your ability to restore them.  TEST YOUR BACKUPS.  You should be able to take bare metal and bring it online at full capacity from nothing more than what you have saved.
pma111Author Commented:
really good pointers...
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Cloud Class® Course: Microsoft Windows 7 Basic

This introductory course to Windows 7 environment will teach you about working with the Windows operating system. You will learn about basic functions including start menu; the desktop; managing files, folders, and libraries.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now