RFC 90: Sustainability roadmap #90
Conversation
thibaudcolas
changed the title
thibaudcolas
changed the title
| | Improvement | Scope clarity | Impact | Reach | Complexity | Measures | | ||
| | --------------------------------------------------------------------------- | ------------- | ------ | ----- | ---------- | ------------------------ | | ||
| | schema.org or similar structured data as a first-party feature | Low | Low | 50% | Low | Time spent on site | | ||
| | Validation of Wagtail sites’ treatment in search knowledge graphs | Low | Low | 50% | Medium | Time spent on site | | ||
| | Yoast-style content analysis & reporting features for Wagtail sites | Low | Low | 50% | Medium | Time spent on site | | ||
| | Support for functionality like Rich Results, Web Stories, etc in CMS | Medium | Low | 5% | Low | Site page views | | ||
| | Better sitemap generation, including more of the site’s contents | Medium | Low | 5% | Low | Server-side energy usage | | ||
| | Redirects awareness, reporting, optimisation to reduce usage where possible | Low | Low | 1% | Low | Server-side energy usage | | ||
| | Canonical page awareness, reporting, optimisation to reduce duplications | Low | Low | 1% | Low | Server-side energy usage | | ||
| | Documentation of front-end performance issues in Wagtail | High | High | 10% | Low | CWV scores | |
There was a problem hiding this comment.
Would it make sense to sort these row by one of the columns (at a glance, I assumed it was sorted by Reach, but the last row is 10%)?
There was a problem hiding this comment.
Good point. I’d probably go for "Impact" and then "Reach" if we did that?
| | Improvement | Scope clarity | Impact | Reach | Complexity | Measures | | ||
| | ---------------------------------------------------------- | ------------- | ------ | ----- | ---------- | ------------------------ | | ||
| | Better CDN support, higher cache hit rates | Medium | Low | 10% | Medium | Server-side energy usage | | ||
| | Official "Wagtail on cloud" hosting guide | Medium | Low | 5% | Low | Server-side energy usage | | ||
| | Support for static sites with 90+% of CMS features | Medium | High | 10% | Medium | Server-side energy usage | | ||
| | Incremental rebuilds for static content | Medium | High | 1% | High | Server-side energy usage | | ||
| | Benchmarking of energy usage - scale-to-zero vs. always-on | High | Medium | 10% | Low | Server-side energy usage | | ||
| | Scale-to-zero CMS database | Medium | High | 10% | Medium | Server-side energy usage | | ||
| | Scale-to-zero CMS server on FaaS or container-as-a-service | Low | Low | 1% | High | Server-side energy usage | | ||
| | Documentation on preview / staging sites’ best practices | Low | Low | 1% | High | Server-side energy usage | | ||
| | Carbon-aware CMS server | Low | Medium | 10% | High | Server-side energy usage | | ||
| | Carbon-aware background jobs | Low | Medium | 10% | High | Server-side energy usage | |
There was a problem hiding this comment.
Same question here, would it make sense to sort by one of the columns?
| | --------------------------------------- | ------------- | ------ | ----- | ---------- | ------------------------------ | | ||
| | Responsive images | High | High | 50% | Medium | Image weight | | ||
| | Optimised Wagtail renditions | High | High | 50% | Medium | Image weight | | ||
| | AVIF images | High | High | 50% | Medium | Image weight | | ||
| | WebP images | High | High | 50% | Medium | Image weight | | ||
| | Lossless image optimisation (PNG) | High | High | 50% | Medium | Image weight | | ||
| | Lossy image optimisation (PNG, MozJPEG) | High | High | 50% | Medium | Image weight | | ||
| | Lazy-loading of images | Medium | High | 50% | Low | Image weight | | ||
| | Optional site images (low-carbon site) | Medium | High | 1% | Low | Image weight | | ||
| | Light YouTube embeds | High | High | 10% | Low | YouTube embed weight reduction | |
There was a problem hiding this comment.
And again... although this might be sorted by complexity?
| This would be unprecedented. | ||
| Net Zero commitments such as the [Paris Agreement](https://unfccc.int/process-and-meetings/the-paris-agreement), [Race to Zero](https://unfccc.int/climate-action/race-to-zero-campaign) or the [SBTi Corporate Net-Zero Standard](https://sciencebasedtargets.org/net-zero) aren't designed for open source projects. | ||
| They’re great ways to drive ambitious climate action nonetheless. |
There was a problem hiding this comment.
I feel like there are implications from calculating Wagtail's carbon footprint on this one that might be worth noting for discussion. If we're including that in the Wagtail project's overall carbon footprint, I think there's some imprecision in any claims to net zero and risk that decisions made wholly independent of the Wagtail project could impact a net zero target.
There was a problem hiding this comment.
Could you expand @willbarton, on the imprecisions in particular? From my understanding you’re right, but this is the case with all carbon footprint accounting / Net Zero claims and wouldn’t be specific to Wagtail or open source more widely.
| The [Web Sustainability Guidelines](https://w3c.github.io/sustyweb/) are a set of guidelines to help organizations reduce the environmental impact of their websites. | ||
| They are currently authored by a _Community Group_ at W3C, and as such don’t have the authoritativeness of web standards. | ||
| They’re still in draft, and not (yet) widely adopted. |
There was a problem hiding this comment.
Would this be controversial? Are there reasons not to follow the WSG (and possibly influence their development by attempting to put them into practice)?
There was a problem hiding this comment.
I see three reasons:
- They’re very recent, and as such harder to make use of than more established sets of guidelines.
- They’re evolving fast, so there’s a chance we’d do things and have some re-work if the guidelines evolve (not super concerned about this)
- They encompass a wider definition of sustainability than what we’re currently aiming for here (see the last sentence below).
There was a problem hiding this comment.
Added a few passing thoughts. This is incredible @thibaudcolas !
| | Better sitemap generation, including more of the site’s contents | Medium | Low | 5% | Low | Server-side energy usage | | ||
| | Redirects awareness, reporting, optimisation to reduce usage where possible | Low | Low | 1% | Low | Server-side energy usage | | ||
| | Canonical page awareness, reporting, optimisation to reduce duplications | Low | Low | 1% | Low | Server-side energy usage | | ||
| | Documentation of front-end performance issues in Wagtail | High | High | 10% | Low | CWV scores | |
There was a problem hiding this comment.
Would we consider our CSP roadmap item here. I know it's security related but from my understanding it helps browsers know when to cache things better and stops additional unexpected scripts from running in the browser. When combined with nonce usage, we get a nice set of improvements for both security and performance in the browser.
E.g. https://techdocs.akamai.com/edgeworkers/docs/content-security-policy
There was a problem hiding this comment.
Good point, I don’t think I’ve seen anyone make that case but there’s probably some performance benefits
| - [Granian – A Rust HTTP server for Python applications](https://github.com/emmett-framework/granian) | ||
|
|
||
| ## Open questions | ||
|
|
There was a problem hiding this comment.
Should we consider our community platforms?
Maybe it's a bit too much scope (or not even measurable), but not only do we extensively use these tools but we also create an example for other communities. This might not need to be a comparison for adopting different platforms but it could involve insuring we see if the platforms actually have their own sustainability considerations/policy.
Off the top of my head; Slack, StackOverflow, Zoom (What's new in Wagtail, meetings), GitHub.
There was a problem hiding this comment.
Yep! Definitely. Will try to think of where to add this.
| | Documentation on preview / staging sites’ best practices | Low | Low | 1% | High | Server-side energy usage | | ||
| | Carbon-aware CMS server | Low | Medium | 10% | High | Server-side energy usage | | ||
| | Carbon-aware background jobs | Low | Medium | 10% | High | Server-side energy usage | | ||
|
|
There was a problem hiding this comment.
Would we also consider the cloud impact of Wagtail itself (our main website, docs hosting) and the CI runners?
There was a problem hiding this comment.
Definitely! I’ll try to think of where to add this. I’m not sure this type of consideration needs an RFC as much, but it’s good to consider.
| | Template caching for admin | High | High | 100% | Low | Server-side energy usage | | ||
| | Wagtail-aware Django debug toolbar | High | High | 5% | Low | Server-side energy usage | | ||
| | SQL performance optimisations | High | High | 50% | Medium | Server-side energy usage | | ||
| | Rewrite performance-sensitive code in Rust | Medium | Medium | 100% | Medium | Server-side energy usage | |
There was a problem hiding this comment.
Would we also consider the CSS and JS assets loading in the admin?
These may not impact the front facing parts of the website but will impact the energy usage of those accessing the admin and the servers that host the Wagtail instances.
There was a problem hiding this comment.
Yes we should! I’ll add it. I’ve been trying to quantify how much usage the admin gets based on internal data about Torchbox websites. It’s really hard.
Co-authored-by: Will Barton <[email protected]>
View the roadmap in HTML. Share this roadmap with a permanent link: wagtail.org/sustainability/sustainability-roadmap/
This addresses the two components of the Sustainability roadmap item from the Wagtail roadmap in preparation for the v6.0 release in February:
There is a bit more to do on sharing our review of WSG and adding items within this document, but the high-level information is there.
Reviewing this
For people reviewing this – there are a few things I would recommend:
I’m also very interested in feedback on the process here, as this is all pretty new to us and there aren’t many projects with this kind of sustainability angle, nor Wagtail contributors with this expertise.
More information