optimism is true moral courage co-founder Science Works @sci_works • metascience cheerleader 🪩 in hock to the sublime
Joined May 2009
- Tweets907
- Following2.2K
- Followers1.5K
- Likes13.9K
Laura Ryan retweeted
The trick-or-treaters have begun arriving. One small witch demanded Haribo. I offered her a marron glacé. Neither of us concealed our disappointment.
Laura Ryan retweeted
You must understand that the think tanker is a kind of artist, and the think tank is a kind of art. One does not simply publish a white paper so much as enter into conversation with the tradition of the medium. The performance must seek to expand the possibilities of the genre.
Laura Ryan retweeted
The new Data Centre Covenant rightly tries to tackle the UK's grid connection problems. But it will make it more expensive to build and could well backfire.
Laura Ryan retweeted
“We always knew planning and grid capacity were problems; this report shows that demand for UK compute might be low too.
If the new Data Centre Covenant piles more costs onto developers who haven’t yet found tenants, it could delay the build-out even more”
cityam.com/healeys-grid-pled… @willecstone
Laura Ryan retweeted
Data centres are a critical part of the UK's future AI play. All signs are that we need to build like mad. Government is targeting +6GW of compute, and the pipeline of planned projects – if built – would exceed that. This is a 4× jump, requiring a 7× speedup from our historic build rate.
It would be an enormous construction feat for a country that has not been that great at building things of late.
So how have we been getting on?
A new report by Henning Zschietzschmann for @sci_works (with @willecstone) shows the picture (thread below). My summary: not great tbh. Only 13% of the pipeline is under construction – most lack planning & power. 42% looks speculative – land developers without tenants, and some sites that DO have planning & power are lying empty. We need hyperscalers to build out the bulk, but big name projects are falling through.
Where are the tenants that we need? Despite the size of the planned pipeline, do we have a problem when it comes to attracting companies to actually build and run the infrastructure?
And what about government policy? Have AI Growth Zones helped accelerate the buildout? Answer: no. First, they only cover 1/7th of the pipeline. Second, projects in Zones have not progressed faster than those outside them.
The government has now announced a new Data Centre Covenant, promising support in exchange for DCs meeting new tests such as ensuring community involvement, hiring apprentices, and buying British hardware. These are good aims – but pile extra burden on developers at a point where projects are already struggling through a slow system. It is the exemplification of @jo3hill's 'everythingism' diagnosis.
Will the new Ministerial Control Centre give projects enough of a boost to offset these costs? Time will tell - but perhaps more importantly, this looks to me like a clear change of direction by the government… perhaps aiming to address the stirring NIMBYism that is increasingly defining the whole debate.
Time will tell whether the policy will help or hinder development. But given what's at stake, it seems we have our work cut out if we are going to have a meaningful stake in this critical ingredient of a more AI-enabled economy and society.
Is there low demand for UK-based compute?
We have an ambitious pipline, but price & planning are bottlenecks, so "government should take extra care to avoid increasing costs to developers."
per yesterday's Covenant:
Strategic data centres must meet the following criteria to qualify for additional government action:
- invest in local community
- create good, well-paid jobs that last beyond construction
- prioritise UK suppliers and technologies and make a meaningful contribution to the development and deployment of British AI hardware
- help build capability in strategically important supply chains
- protect nature, minimise water use and publish annual energy and water consumption reports
- capture the heat they produce for nearby homes, schools, hospitals or businesses wherever practical
- & more
If your project doesn't, your grid connection might be reallocated to one that does. But it's optional!
🙃
Laura Ryan retweeted
This @sci_works report is an astonishing contribution, tracking the reality of UK’s datacentre buildout
We need to both upgrade our ambition and massively improve throughput
Laura Ryan retweeted
when governments engage in the Pravda-esque recitation of big numbers to signal progress, it's worth asking what they're not telling us – important work here
Laura Ryan retweeted
We always knew planning and the grid held back data centres. Our new report finds a third problem: 42% of the 2030 pipeline has no named tenant.
Schemes to help the build-out need to fix the grid and planning without making building here too expensive.
Laura Ryan retweeted
We have a new Data Centre Covenant. But, until now, there was no detailed overview of what data centre capacity the UK plans to build: how much, where, and when?
What signals are there about demand for UK compute?
Our new report has the answers – which may surprise you! 🧵👇
Laura Ryan retweeted
I sometimes worry that, as scientists, we get caught up in the sugar rush of being able to run a million experiments at once (or at least clear through our backlog of ideas to get signal) and subtly lose our taste for judging which problems are worth working on in the first place
Laura Ryan retweeted
Frances Macdonald ( 1873 – 1921), prominent Glasgow School artist, whose design work was a huge feature of the British Art Nouveau style from the 1890s until the early 20th century #WomensArt #GlasgowGirlsWeek
Laura Ryan retweeted
An exceptional opportunity to partner with leading thinkers in AI and scientific discovery and fix what's broken about scientific software!
Looking forward to discussing (and judging) the myriad creative projects that this diverse group will build.
Tomorrow! 🗓️
Our tools for science hack with @inherent_labs is hugely oversubscribed, with half of hackers coming from outside the golden triangle, thanks to @ukdynamism 🇬🇧
Some inspiration - the tools researchers flagged as bottlenecks👇
Can we build some solutions? 🧪
Laura Ryan retweeted
Tomorrow! 🗓️
Our tools for science hack with @inherent_labs is hugely oversubscribed, with half of hackers coming from outside the golden triangle, thanks to @ukdynamism 🇬🇧
Some inspiration - the tools researchers flagged as bottlenecks👇
Can we build some solutions? 🧪
London AI × Science Hackathon 👾
Science Works × @inherent_labs
• Oct 16-18
Lots of scientific software is outdated, hard to use, and isn't interoperable – let alone making use of (or ready to interface with) AI.
Or it's expensive. Join us to build better for the community
Laura Ryan retweeted
Look at him, thats my Qual.
Your what?
My Qualitative! My vibe specialist. Look at him, you notice anything different about him? Look at his face.
I'll give you a hint, his name is Jayden. He won a local poetry competition in bushwick! Yeah I'm sure of the vibe.
Quant firms are hiring idea guys, not quants
I've been skeptical about the whole - software is commoditized, ideas are the moat argument
But I've seen first hand the best quant and technical firms in the world hire thinkers, liberal arts grads, and generalists
I predict the trend accelerates
Letter from @AQRCapital below
Laura Ryan retweeted
Unfortunately, while the after may *look* nicer to you and me, it has been found to be a pastiche. Work to return it to its authentic state is under way.
Putting a facade on a prefab concrete block in the former east Berlin. The underlying structure is unaltered; instead, the architects have:
- Given it a coat of plaster;
- Painted it in varying colours;
- Broken up the massing with some clever tricks to the window and roofline detailing.
Look closely: the floor height remains continuous and the windows are all in the same places as before. The effect of spontaneity and variety is created by using ornament create patterns of group and emphasis.
Laura Ryan retweeted
Much of scientific research runs on old software with clunky interfaces or unexplained bugs. To realise the benefits of AI for science, we first need to fix the foundations.
Join us to pay off science’s tech debt. Tokens, travel and accommodation provided, prizes to follow...!
London AI × Science Hackathon 👾
Science Works × @inherent_labs
• Oct 16-18
Lots of scientific software is outdated, hard to use, and isn't interoperable – let alone making use of (or ready to interface with) AI.
Or it's expensive. Join us to build better for the community
wave upon wave of validation as a sixth researcher lists ImageJ as a tool they want to rebuild
London AI × Science Hackathon 👾
Science Works × @inherent_labs
• Oct 16-18
Lots of scientific software is outdated, hard to use, and isn't interoperable – let alone making use of (or ready to interface with) AI.
Or it's expensive. Join us to build better for the community
We're exceptionally pleased that – thanks to @ukdynamism – we can now offer travel & accommodation grants for prospective hackathon participants outside the golden triangle. 🇬🇧
Scientists have software problems to solve across the country – help us get some in the room 📩 📩 📩
London AI × Science Hackathon 👾
Science Works × @inherent_labs
• Oct 16-18
Lots of scientific software is outdated, hard to use, and isn't interoperable – let alone making use of (or ready to interface with) AI.
Or it's expensive. Join us to build better for the community