Fershad Irani - Notes

Monthly #23

2025-12-19T00:00:00Z

Lol these monthly posts are consistently getting done later in the month. At least that's consistent, right?

November 2025

🇯🇵 TPAC 2025 in Kobe, Japan. Having spent 3 weeks in Japan through October, I was back again in the second week of November for the W3C's annual TPAC gathering. It was a terrific experience, though was completely draining at the same time. Besides all the peopleing at the conference itself, I had a pretty decent shopping list from Kay of all the things we'd held off on buying on our holiday to Japan which I was supposed to pick up this time around. I knew the supermarkets & shopping streets of Kobe's Sannomiya area pretty well by the end of the week.

I've written more about my time at TPAC in Reflections on TPAC 2025.

🏉 Coaching Youth Touch in Taipei. November also saw the start of the Touch season at the local American school here in Taipei. The girls at the school play Touch, while the boys play Rugby from November through to the end of January. This year I'm back coaching the Junior Varsity team, a challenging task given the wide skill range of the players in the squad, but one that is always rewarding as players develop through the course of the short season.

This year, for the first time, our team also has an end of season tournament which we are training towards! Exciting stuff for all involved.

Tracking start, stop, continue targets

These are my current Start, stop, continue items for 2025. I've made one change to the "Start", removing drinking a glass of water every morning (that's become habitual) and replaced it with a goal to get 2 days at the gym in each week for the rest of the year.

Start

Tracking habits daily: 😔 Nope. I'm canning this one, it's not going to happen.
Carrying a water bottle with me: 😔 Not really, as the temperatures have gone down a bit I don't feel the urge to hydrate while I'm out.
2 days at the gym each week: 😅 No, it's just not possible with coaching time being squeezed in around Green Web Foundation work.
Making a weekly "Links & Listens" list: 😅 Haven't updated the links page in a while, sorry.
Cooking more regularly: 😔 Nope.

Stop

Reaching for sugary (fizzy) drinks: 😔 Nope. Japan. Convenience stores.
Not waiting for things to be perfect before releasing them:😅 Nope. Working on some major changes to our verification process at work, which is probably something we'd want to have close to finalised before putting it out there~

Continue

Trying to get Touch into more local schools in Taiwan: 😔 Nope. Things have been quiet on this front.
Getting in a 20-30 minute walk during the day: 😄 Yes! Much walking around Kobe especially.

Let's get Power Profiling into Chromium DevTools

2025-11-25T00:00:00Z

This morning I wrote about how back in 2014/15 we almost had power profiling land in the Chromium DevTools.

Tonight, I've created a new ~~issue~~ feature request on crbug.com with the hope that it can get the ball rolling on getting power profiling into the Chromium DevTools within some timeframe that isn't tomorrow but also isn't never. Check it out at [Feature request] Power profiling to surface power and energy draw from browser and website use.

Help this get traction

If you care about this and think that power profiling should be in the Chromium DevTools, please visit the issue using the link above and give it a +1 (screenshot below). This help show that there is a desire for a broader community this kind of functionality to be part of DevTools.

Use the +1 button on the crbug issue page to show that having power profiling in Chromium is important for you.

Hopefully we're on the way to getting power profiling in multiple browsers. 🙏🏼

Monthly #22

2025-11-20T00:00:00Z

A suppppper late monthly this time around due to a very busy first few weeks of November squeezed in around a few bouts of sickness.

October 2025

🇯🇵 3 amazing weeks in Japan. For Kay and I, October was our big break month in 2025. We ventured to Japan, partly in order to meet up with my sister and her husband who were traveling there at around the same time. We met them for a day in Kyoto squeezed in between spending time in Hiroshima, Kurashiki, Kinosaki Onsen, Matsue, and Izumo.

It was a wonderful three weeks, visiting a few spots for the first time while also exploring some places we'd already visited a little deeper.

A definite highlight were the 6 days we spent in Matsue and Izumo, a part of Japan that neither Kay or I had been near before. Matsue in particular was such a chill city, with things operating at a pleasant pace, and some beautiful spots along the lakeside to spend the afternoon. We happened to be there in time to catch the end of the city's lantern festival around Matsue Castle, which included a really nice little open air night market.

I'd love to explore that region a bit more if time allows ~~

Tracking start, stop, continue targets

Start

Tracking habits daily: 😔 Nope. I'm canning this one, it's not going to happen.
Carrying a water bottle with me: 😄 Yes!
2 days at the gym each week: 😅 No, but I did get in a run every 3 days in Japan. Managed to get my pace down to 5:29/km which is pretty solid for me historically.
Making a weekly "Links & Listens" list: 😅 Haven't updated the links page in a while, sorry.
Cooking more regularly: 😔 Nope.

Stop

Reaching for sugary (fizzy) drinks: 😔 Nope. Japan's got some wild sodas in the convenience stores.
Not waiting for things to be perfect before releasing them:😅 Nope. Working on some major changes to our verification process at work, which is probably something we'd want to have close to perfect before putting it out there~

Continue

Trying to get Touch into more local schools in Taiwan: 😔 Nope. Things have been quiet on this front.
Getting in a 20-30 minute walk during the day: 😄 Yes! Much walking in Japan. Many sightseeing.

Meet me at TPAC 2025

2025-10-23T00:00:00Z

I'm really excited to be heading to Kobe, Japan next month for the W3C's annual TPAC conference. I'll be there as an individual invited expert, and am looking forwards to meeting some of the folks working in the Sustainable Web Interest Group in person. I'm also hopeful that my introvertedness won't get in the way of me having conversations with folks working across other W3C working groups, interest groups, and community groups who's knowledge might help unlock some of the projects we've got bubbling away at the Green Web Foundation.

If you'll be in Kobe from November 10 - 14 let me know, it'd be great to meet up!

Monthly #21

2025-10-10T00:00:00Z

September was a month for things to ramp down and other things to ramp up on the work front. Away from that, it was a pretty quiet month.

September 2025

👨🏼‍💻 As GAW ends, other stuff begins. The Grid-aware Websites (GAW) project has been the center of my attention at work for the last 12 months. I'm so proud of being able to build out the code library, deploy it to the Branch website, and work with Hannah and the outstanding group of individuals that we have in our advisory group. All that said, funding for the project ended in September and so we spent a lot of the month wrapping it up. There'll be a project reflection blog post that should come out sometime later this month which I'll share here too.

As GAW wrapped up, I shifted my attention to what will probably be my big project for the first half of 2026 which is to update the verification criteria we have at the Green Web Foundation. There'll be more about that on the Green Web Foundation blog in November so stay tuned over there.

🏉 Touch goals for 2026. There's a few international Touch tournaments coming up in 2026, and I'm still pretty keen to run around representing Taiwan as a player in those. As the body gets older though, it means that I need to start the preperation work a bit earlier to keep up with the pace of the game. The main aim is an Opens division tournament in Australia which I'm hoping to be selected for. With an eye to that, I've started getting in regular weekly gym sessions especially to build back the strength in my legs. October will be the month I begin supplementing that with some actual running, which I loathe.

Tracking start, stop, continue targets

Start

Tracking habits daily: 😔 Nope. I'm canning this one, it's not going to happen.
Carrying a water bottle with me: 😄 Yes!
2 days at the gym each week: 😅 Kinda. For two weeks, yes and for another two weeks I only managed one session.
Making a weekly "Links & Listens" list: 😅 Kinda. Not weekly, but I've got a links page which I'm now updating infrequently.
Cooking more regularly: 😔 Nope.

Stop

Reaching for sugary (fizzy) drinks: 😔 Kinda. It's hard during a brutal Taiwan summer with 7-11 around every corner!
Not waiting for things to be perfect before releasing them:😅 Kinda. I'm sitting on many project/writing ideas just because the time isn't there to do them.

Continue

Trying to get Touch into more local schools in Taiwan: 😔 Nope. Things have been quiet on this front.
Getting in a 20-30 minute walk during the day: 😄 Yes!

How should the Web Sustainability Guidelines be organised?

2025-09-20T00:00:00Z

The first version of the W3C Web Sustainability Guidelines (WSG) were initially published in 2023, under the stewardship of the Web Sustainability Community Group. Since then, a charter has been established for the formation of the W3C Web Sustainability Interest Group (WSIG) in 2024, and that group has been refining and editing the WSG for the past year.

The status of the WSG is that they have gone through a thorough editing process, and the Editor's Draft is now beginning the early stages of review and feedback from other groups in the W3C. This is an important step towards eventually releasing the next version of the WSG.

I've been involved in the group as an Invited Expert, but mostly my engagement has been asynchronous largely due to the timing of the regular weekly meeting. That said, all those meetings are recorded and shared on GitHub which allows me to catch up with things when I can.

One of the things that the members of the IG have been discussing over the past few months has been how to best organise the WSG. There's a few things to note here:

There are 92 guidelines at the moment (with 254 associated success criteria).
They have a very broad scope, covering everything from design, development, through to business strategy.
The guidelines currently have a roles based organisation (UX, Web Dev, Hosting & Business).
There is a additional level of tagging for guidelines which allows for folks to apply filtering themselves.

The sheer number of guidelines makes it impractical to just put them in a single list. The broad scope also means that there are a number of guidelines at the moment which have potential overlap (something that there is an open issue to work through). Within the group, there are also reservations about the effectiveness of the role based organisation, as it might mean that people who come to the WSG for the first time don't see a relatable role listed and bail.

Request for feedback

This all leads to the key question the WSIG is grappling with at the moment - how to organise such a varied set of guidelines to make them usefully navigable for people who are looking into web sustainability (perhaps for the first time)?

I've jotted down some of my thoughts in this open GitHub issue, and the topic has been discussed in detail across several WSIG meetings. Getting input from outside the group would certainly be extremely beneficial for everyone involved.

So let us know what you think. Does the current roles based approach work for you? If not, what adjustments could we make so that the guidelines are presented in a manner that are navigable, align with the world they'll be applied in, and can help encourage broader adoption?

All ideas are welcome, and there are no wrong answers. You should be able to leave a comment on GitHub issue I've linked to above, or if you'd rather share stuff privately then feel free to reach out using any of the platforms I've listed on my /contact page. Any feedback is welcome and appreciated 🙏🏼.

Monthly #20

2025-09-07T00:00:00Z

August was a rare month for 2025, one where I was at home for the entire month!

August 2025

🍰 38 trips around the sun. The highlight of August, as it is every year, was my birthday. I normally don't do too much to celebrate. This year, Kay and I went to a new vegan restaurant that has opened near home to try it out.

🏉 3 week Touch tournament. I organised a 3 week Touch tournament during the month. Since it's Summer here the temperatures during the day can be pretty intense, so we ran a night tournament. It was small, with only three teams registering, but it was great getting folks together playing some regular Touch.

Tracking start, stop, continue targets

Start

Tracking habits daily: 😔 Nope.
Carrying a water bottle with me: 😄 Yes!
2 days at the gym each week: 😔 Not yet. Started in September ~~
Making a weekly "Links & Listens" list: 😔 Kinda. Not weekly, but I've got a links page which I'm now updating infrequently.
Cooking more regularly: 😔 Nope.

Stop

Reaching for sugary (fizzy) drinks: 😔 Kinda. It's hard during a brutal Taiwan summer with 7-11 around every corner!
Not waiting for things to be perfect before releasing them: 😄 Yes! Many a blog post have been published this month!

Continue

Trying to get Touch into more local schools in Taiwan: 😔 Nope. Things feel through with a local school in central Taiwan. Hopefully there'll be an opportunity later this year.
Getting in a 20-30 minute walk during the day: 😄 Yes!

On Google's AI impact paper

2025-08-26T00:00:00Z

As with everything on this website, opinions expressed are my own. Also, this might ramble. I'm sorry.

A recent paper published by a number of Google staff shared some data on the energy and water usage from using their Gemini AI models, as well as carbon emissions information. The paper is here if you want to read it https://arxiv.org/abs/2508.15734.

Cue polarised debate on LinkedIn and other parts of the web that either:

This report shows that claims of the negative environmental impacts of AI are overblown. Or,
This report lacks data and transparency to make the data actually useful.

David Mytton is in the first camp (environmental impacts overblown), and has written a summary on Substack. Ketan Joshi is in the second camp (lacks transparency) has written a summary on his blog as well.

Here's my summary:

Google employees release paper that paints their employer's AI platform in a positive light.

Like it's great that someone at Google thought about putting some resources towards this. But come on guys, to make this meaningful open up the data so that it's available for independent researchers to explore.

Both Ketan and David have gone into the shortcomings of the report's use of median values, and the fact that only data for single text queries was analysed. To be fair though, single text queries might be a suitable lens to use for an analysis of Google's AI system - because that's the kind of query that is triggering the unsolicited AI Summaries that are being forced upon users with almost every search result now. Maybe there just wasn't enough data for the other Gemini applications like video or code generation because people just aren't using those as much. Probably since those features aren't being being forced upon them. Yet.

Is it bad for the environment, or not?

Ask this question on LinkedIn. Go on, I dare you.

The way I think about it is - at the level of an individual text query, sure it might seem close to insignificant. But the scale of hundreds of millions of queries per day, that impact probably becomes meaningful.

(Aside: I think the same way about website sustainability.)

That's just thinking about a single text query. Unfortunately this paper doesn't really give us any real detail about scale (despite the title). Nor does it allow us to think about multiple AI use cases.

One environmental argument I definitely don't buy is that we'll efficiency our way out of this as David suggests at the end of his blog post:

There is a huge buildout of AI infrastructure under way and it’s likely that total energy consumption related to AI is going to grow. However, I expect efficiency gains, batching, and load‑placement should temper growth versus extrapolations.

These facilities, which are being built today, have an immediate and real impact on the physical world around them. That's felt in the air, the water, the land, and also the energy grid. No amount of efficiency improvement can restore that. Buildout, sure. But buildout within bounds.

Sceptical about website carbon emission figures

2025-08-14T00:00:00Z

ahem, has anyone calculated how much carbon their Storybook uses when visited … I'm guessing it could be pretty chunky.
— Stuart Robson (@sturobson.com) Aug 13, 2025 at 20:01

I’m deeply suspicious of anything that tries to estimate how many kgs of carbon a website ‘emits’
There are so many factors and we don’t have strong data
— Andy Davies (@andydavies.me) Aug 13, 2025 at 21:00

Stuart Robson asked the question above on Bluesky, to which Andy Roberts was one of a few respondents. I've been tagged in that thread, and rather than writing a multi-part reply on Bluesky I'll capture my comments as a note here because it feels like something that will be handy to refer back to in the future.

Reply

Being sceptical about website carbon emission is fair enough. As Andy's pointed out there are a lot of moving parts and limited data points. But there's an environmental impact from our web activity which is what these website emissions calculations aim to surface. So let's talk about what those emissions figures represent, and what alternatives we have.

Estimating

Most of the website emission figures you'll see online are estimates. We're not talking about values that have been reached through measurement (more on that in a bit). The most widely used website carbon estimation model is the Sustainable Web Design Model. This model is attributional, and at a very high level it works like this:

(A) Take a value for the total energy used by the internet
(B) Take a value for the total data transferred over the internet
Divide A by B to arrive at an amount of energy used to transfer a unit of data.
Multiply that by the amount of web page data transferred to get the estimated energy used.
Multiple that by a carbon emissions intensity value to get the estimated carbon emissions.

The latest version of the Sustainable Web Design Model gets a bit more detailed than this (see the methodology), but at a high level that's what it's doing. Emissions results can be broken down by system segment (backend, network, frontend) and operational/embodied phases.

This means two things:

Data transfer is used as a proxy for website emissions. This isn't great (see my thoughts here, and here) but it is the best thing (most accessible data point) we have available to us absent of more complex tooling.
Due to the attributional (top down) nature of the model, the resulting emissions values is likely to be an over-estimate.

All that said, it's a tool in our toolkit and a good first step to get folks into the ballpark. Understanding how it works can help better interpret the results. The ethos behind models like this is progress over perfection.

Measuring

That gets us to asking what "measurement" can we do instead to get better data?

It is absolutely possible to measure the energy used by a web page. It just takes more tooling, or a dedicated setup, which makes it not so easily approachable for most folks. There just aren't the browser/web APIs available at the moment to make this something that can be shown up in DevTools.

Firefox Profiler is the only browser profiling tool that I'm aware of which surfaces power and energy measurements in profiles. Just like with web performance though, reading profiles can feel intimidating to most regular folks and needs some level of knowledge to interpret. The power profiles captured in Firefox also differ in how much detail they show based on the underlying system architecture, and only capture operational energy use on the device.

Outside of that, groups like Green Coding Berlin are working on platform solutions that enable measurement. However, these require a dedicated setup/additional configuration which again adds a barrier to adoption.

All that said, when possible, "measure" rather than "estimate". And, lean towards talking about "energy use" rather than "carbon emissions" especially for the frontend (see this recent post I wrote on that topic).

What is the goal of the Web Sustainability Guidelines?

2025-08-12T00:00:00Z

Łukasz raised this question in the Web Sustainability Guidelines GitHub repository a couple of months ago as part of understanding and setting a goal for the measurability taskforce with the Sustainable Web Interest Group:

There's a number of different paths to progress the work in the task force and in order to decide which of them to take we'd like to define ultimate goal of the Sustainable Web Design guidelines (followed by a definition of measurability task force goal).

It's come up again, and has been flagged for discussion in the group call this week. I'm capturing this note with my reply which I've also shared on GitHub.

This issue was flagged by @TzviyaSiegman as a conversation topic for the next group call on August 14. I'll leave my thoughts here as I'm unable to attend that call.

I advocate for adopting a narrow scope when goal setting for the WSG. Throughout W3C writing and documents there are constant references back to "the web", and I believe that we that we should keep that central to how we frame the mission of the WSG as well.

Centering a framing around "the web" allows for two things:

It allows for a narrower target audience to be defined.

It reduces the complexity of the guidelines.

Both of these go hand in hand. Having a defined target audience will greatly help with how guidelines and supplemental material are thought about, selected, and written. It's also an important part in being able to clearly communicate the guidelines.

For reference, here's the target audience for the WSG from the current introduction:

Web Sustainability Guidelines promote planetary, people, and prosperity best practices based on measurable, evidence-based research; aimed at end-users, stakeholders, website or application creators, tool authors, tool authors, educators, students, policymakers, purchasing agents, product owners, managers, and decision-makers

That's a broad audience, and you'd speak to different groups from that list in different ways.

Other frameworks and certifications (e.g. B Corp, RE100) already exist that address sustainability at the organisational level. It just feels beyond the scope of W3C guidelines to be getting involved at that level. Rather these other initiatives can be signposted to like "hey if you're looking to adopt these web sustainability guidelines, then your organisation would also benefit from these other initiatives". Individual guidelines could even link out to these where appropriate.

When thinking about "the web" in the context of writing guidelines, another question emerges - who are we writing for? Those who are building on the web (designers, developers etc), or those who are building the web platform (browser engineers)? For mine, I think the WSG are designed the former group (building on the web) at this stage and should be compiled with them as the primary audience.

So with all that said, here's a (long) one liner from me defining the goal of the WSG:

The goal of the Web Sustainability Guidelines (WSG) is to define strategies and practices for those building on the web to make web content, online products, and digital services more sustainable through the lens of planet (environment), people (individuals, community, society), and prosperity (shared equity).

Got thoughts? Let me know, or contribute over on GitHub.