Hacker News: twak

New comment by twak in "It is time to give up the dualism introduced by the debate on consciousness"

twak — Mon, 18 May 2026 09:48:13 +0000

I've yet to find a falsifiable definition of consciousness.

I do believe in intelligence (which is measured against a particular task) and ego (which inflates the self over the other).

New comment by twak in "Opus 4.5 is not the normal AI agent experience that I have had thus far"

twak — Wed, 07 Jan 2026 14:59:44 +0000

you can build it and simply use it in your own office? There is no need to shout about it if the cost of writing software goes to zero (but the value remains non-zero!).

New comment by twak in "AI: Accelerated Incompetence"

twak — Wed, 28 May 2025 14:01:00 +0000

i agree; but perhaps also it is the difference between managers and SWE? The former (SWE team leaders included) can see that engineers aren't perfect. The latter are often highly focused on determinism (this works/doesn't) and struggle with conflicting goals.

Through a career SWEs start rigid and overly focused on the immediate problem and become flexible/error-tolerant[1] as they become system (mechanical or meat) managers. this maps to an observation that managers like AI solutions - because they compare favourably to the new hire - and because they have the context to make this observation.

[1] https://grugbrain.dev/#:~:text=grug%20note%20humourous%20gra...

New comment by twak in "An online exhibition of pretty software bugs"

twak — Sun, 11 May 2025 20:59:41 +0000

i used to do lots of graphics, and love these bugs. They give an enjoyable insight to the development process and algorithms as the progress through a project!

i have my own archive of my own bugs/"artworks" - https://twak.org/glitches-in-the-worlds-geometry-engine/

New comment by twak in "Project Aardvark: reimagining AI weather prediction"

twak — Fri, 04 Apr 2025 11:48:04 +0000

The Turing had an interesting approach to naming, not only stealing the Glasgow group's name, but also choosing the initials 'ATI' (in 2015...).

It's recently struggling for relevance.

https://www.ft.com/content/6bfea441-e16c-499a-a887-69f735c29... (https://archive.ph/ujfhb)

I hope they turn it around because the UK need for AI academic coordination/leadership is so high.

New comment by twak in "Aicracy – Governed by Algorithms"

twak — Mon, 30 Jan 2023 12:27:39 +0000

My old supervisor wrote a book on computational socialism back in 93. I rather like some of the ideas (can a participatory centrally planned economy work now that we have internet/cybernetics?), but suspect it wouldn't work in practice.

https://en.wikipedia.org/wiki/Towards_a_New_Socialism

New comment by twak in "Is DALL-E 2 ‘gluing things together’ without understanding their relationships?"

twak — Thu, 04 Aug 2022 16:35:10 +0000

fantastic book. Made me consider the question of whether consciousness exists at all or if it is just some hack by evolution to allow introspection.

I haven't found a definition of consciousness which is quantifiable or stands up to serious rigour. If it can't be measured and isn't necessary for intelligence, perhaps there is no magic cut-off between the likes of Dall-E and human intelligence. Perhaps the Chinese-room is as conscious as a human (and a brick)?

New comment by twak in "Deep Learning Papers Are Kinda Bullsh-T"

twak — Tue, 19 Jul 2022 17:16:26 +0000

Other causes include the pressure to publish quickly in ML (while your approach is en vogue), with small teams, before your funding runs out, while hitting conference deadlines.

In these situations, I have suggested releasing anonymous implementations after the paper is accepted just to get the code out there. I am not certain this is the right thing to do!

New comment by twak in "Deep Learning Papers Are Kinda Bullsh-T"

twak — Tue, 19 Jul 2022 14:39:10 +0000

Academics are judged by the publications not their implementations, so the system favours over-sold manuscripts and it-ran-once implementations. Until funding is conditional-on (and provided for) robust well maintained code it will remain challenging to get reproducibility.

Frequently the PIs (bosses) will not even glance at the repositories written by junior members, probably can't read code anyway, and certainly won't allocate time for their maintenance. Even worse, most academics who do publish code have never been exposed to real world software engineers, their techniques, or tools.

New comment by twak in "Deepfake used to attack activist couple shows new disinformation frontier"

twak — Wed, 15 Jul 2020 13:49:37 +0000

> It’s not to say the developers and scientists at Adobe are lesser

adobe research scientists are crazy-strong in the area of deep/neural graphics[1]. Perhaps we should disentangle adobe research from photoshop?

[1] https://research.adobe.com/publications/

New comment by twak in "Self-hosted, super simple photo stream"

twak — Wed, 11 Mar 2020 13:03:22 +0000

Look awesome - I've been looking for some hackable code like this to present my photos from google drive. Cheap photo hosting from google + minimal load on my server...

New comment by twak in "The cyclist behind an anti-cyclist Facebook group"

twak — Fri, 10 Aug 2018 14:32:59 +0000

Within the UK there is a large variety of cycling cultures. Cambridge is a city where everyone cycles to work in jeans/office clothes.

(Don't confuse "track bikes" and "road bikes"...)

New comment by twak in "The cyclist behind an anti-cyclist Facebook group"

twak — Fri, 10 Aug 2018 14:25:04 +0000

a counter point: when cycling in London you have to be aware that 30% of the time there will be be a vehicle stopped the protected bike zone in front of the lights...

Show HN: Defeating the Snooper's Charter by Visiting Bad Sites All the Time

twak — Sun, 20 Nov 2016 01:04:26 +0000

Article URL: https://github.com/twak/hot-fuzz

Comments URL: https://news.ycombinator.com/item?id=12996681

Points: 10

# Comments: 1

New comment by twak in "Building better video"

twak — Sat, 13 Aug 2016 23:15:37 +0000

I guess I was thinking that a good way to simplify many video UIs would be to remove these buttons. Perhaps the browser-esque way of providing this functionality would be to use bookmarks instead?

Building better video

twak — Sat, 13 Aug 2016 13:49:56 +0000

Article URL: http://www.twak.co.uk/2016/08/building-better-video.html

Comments URL: https://news.ycombinator.com/item?id=12281487

Points: 3

# Comments: 2

New comment by twak in "Show HN: Strobee in-browser video splicing (use Chrome)"

twak — Mon, 30 May 2016 13:54:53 +0000

Goal was to create a platform for sharing user video that would otherwise be too awful to watch. By dropping length, but increasing variety, it's easier to create something like to watch...

More info on why we do this in the browser: http://www.twak.co.uk/2016/04/browser-video-splicing.html

Show HN: Strobee in-browser video splicing (use Chrome)

twak — Mon, 30 May 2016 13:51:41 +0000

Article URL: https://strobee.net/index.html?user=146e5040000000&startAt=146e5040000000_11158600000000

Comments URL: https://news.ycombinator.com/item?id=11801329

Points: 5

# Comments: 1

New comment by twak in "Procedural Modelling of Buildings [pdf]"

twak — Sun, 07 Jun 2015 22:07:50 +0000

The papers written at Esri R&D Zurich this year also include: CGA++ http://research.michael-schwarz.com/publ/2015/cgapp/ Interactive Handles (by me!) http://twak.blogspot.ch/2015/03/interactive-dimensioning-of-...

Pascal has published, and keeps publishing lots of stuff on procedural modeling (all while directing the Esri research center): https://scholar.google.co.uk/citations?user=zzMlobMAAAAJ

New comment by twak in "Procedural Modelling of Buildings [pdf]"

twak — Sun, 07 Jun 2015 22:02:04 +0000

It's always slightly surreal to see your employer on the front page of hn, even more so when you find you find your colleague making the second comment.