dbuxton

One thing I find interesting about discussions of typography in Cyrillic is how poor the overall readability of text is in most fonts compared to Latin because of the relative scarcity of risers and descenders (e.g. pqlt etc)

One of my tutors at university claimed that she was able to read 9th century manuscript Cyrillic faster than modern printed books because the orthography was more varied and easier to scan/speed-read.

(That wasn't something I found to be true)

It’s hard to bet against the foundation models winning consumer use cases where you can reimagine the whole product as a single tool or small number that can be dynamically plugged in to the underlying model and doesn’t require access to proprietary data/custom context.

I think this is a relative succinct summary of the downside case for LLM code generation. I hear a lot of this and as someone who enjoys a well-structured codebase, I have a lot of instinctive sympathy.

However I think we should be thinking harder about how coding will change as LLMs change the economics of writing code: - If the cost of delivering a feature is ~0, what's the point in spending weeks prioritizing it? Maybe Product becomes more like an iterative QA function? - What are the risks that we currently manage through good software engineering practices and what's the actual impact of those risks materializing? For instance, if we expose customer data that's probably pretty existential, but most companies can tolerate a little unplanned downtime (even if they don't enjoy it!). As the economics change, how sustainable is the current cost/benefit equilibrium of high-quality code?

We might not like it but my guess is that in ≤ 5 years actual code is more akin to assembler where sure we might jump in and optimize but we are really just monitoring the test suites and coverage and risks rather than tuning whether or not the same library function is being evolved in a way which gives leverage across the code base.

Interesting that the infographic (which I thought was exceptionally well-designed, well done USGS) found it necessary to call out that 0 billion gallons/day goes to Mexico. Was this done by previous or this administration I wonder? I do recall reading something about disputes between US and Mexico over abstraction. (Presumably from Rio Grande or similar).

I once lived in Moscow on a compound with an adopted rescue dog. The compound had a shock collar and invisible fence setup.

Moscow’s street dogs are renowned for their intelligence. I have seen street dogs taking the escalators on the Metro. This dog worked out not just that the beeping + discomfort was worth the freedom, but also that he could wear out the battery faster by going up to the very edge of the fence - where the chirps became an uninterrupted beeeeep - and as soon as the beeping stopped, whoosh he was gone.

Great question. For me reliability is variance in performance and capability is average performance.

In practice high variance translates on the downside into failure to do basic things that a minimally competent human would basically never get wrong. In agents it's exacerbated by the compounding impact of repeated calls but even for basic workflows it can be annoying.

Fundamentally, we are at a point in time where models are already very capable, but not very reliable.

This is very interesting finding about how to improve capability.

I don't see reliability expressly addressed here, but my assumption is that these alloys will be less rather than more reliable - stronger, but more brittle, to extend the alloy metaphor.

Unfortunately for many if not most B2B use cases this reliability is the primary constraint! Would love to see similar ideas in the reliability space.

I had to do a MN1 application for my US-born daughter as both me and my partner were born abroad. As OP alludes to this is a sort of side quest called "registration" which if you don't do it before the child is 18 lapses (although I think there are still routes to obtain citizenship in these circs).

The most difficult part of the process (not dealt with in this version of Passport Application but maybe a future DLC pack?) was actually finding someone who could certify my evidence (you are meant to submit originals but they keep the docs including passports for 3-6 months which is a bit unrealistic if you are living abroad). I can't remember the exact rules but it wasn't possible to use a US notary or a normal solicitor certification process and instead I needed to go to a council office.

After calling about 5 councils all of whom disavowed any knowledge of the process or its requirements I ended up finding someone at Islington Council who was delightfully helpful. But it was one of the more frustrating UK government interactions I've had.

Not caused but correlated: doing something risky is much easier when you have something to fall back on - which I’d guess is a lot more common in the Ivies

For me this is great for practice (I tried Russian). However the big missing piece for all these language learning apps is the lack of support for spotting and correcting errors in your pronunciation - as long as you say the word more or less right, the transcription gives you a pass.

I am very excited for the whole STT/TTS to go away and for us to have models that really "hear" exactly what you said.

Sometimes this is about accent but a lot of the time, the AI won't spot areas where you e.g. fudge a case ending or the stress on a word. Yes, you can get some of that pronunciation right by the AI repeating back with the correct stress or clear case, but you never really get the confidence that you would get from an actual human.

Another product suggestion - turn off transcription (at least for the tutor side of the conversation; I'd suggest both). Personally I find it distracting at best for languages I already speak well and a crutch for those I don't.

Finally, I find it really very hard to enjoy having a random conversation that's not very directed ("What interests you most about artificial intelligence?"). I'd suggest that there are ways of making it more goal focused without being explicitly gamified - maybe something like, here's a position and you have to persuade me (AI debate club!), or something that brings out an actual opinion or relates to a concrete experience ("what's your main goal in your job this year").

Overall though this is the first product I've seen in this space that I might actually use, so well done.

I also got Anton. Looks like something's hard coded - or maybe a caching issue?

I came here assuming that this was a clone of Gemini CLI (which is a clone of Claude Code). Both pleased and disappointed to be wrong.

I find it weird that this is news and not:

- That it's still way cheaper in most instances to book a return (especially where the "trip" straddles a weekend) rather than a one-way fare when travelling long haul - even if you just throw away the return flight.

- That you can sometimes get access to totally different inventory by booking a package including accommodation, even if that accommodation is one night in a shared dormitory in a hostel (which you just don't go to).

At least group discounts have a recognizable economic rationale. But in these examples you are getting a strict superset of the same SKU (OK, maybe the change rules might be a little tighter, but not in a way that's perceptible) for less money.

Although in this case you actually have to be accompanied by another adult

I can't wait for the Rust port of QBasic Gorillas

Are there extensions that e.g. use NLP/LLMs/vectors to suggest potential links from elsewhere in ones KB? Could be a fairly straightforward plugin.

(I haven't used Roam personally and have no idea if it even has a plugin architecture or is extensible, but this reminds me a lot of some of the knowledge management work we're doing with corporates)

Hey! Congrats on the launch. I just signed up for a trial account and I’m pretty impressed with the search API (haven’t used websets yet but looks cool).

Our experimental use case is enabling quick and dirty integration of web-based docs into an employee service agentic chatbot - lots of the questions are around “how do I max out my 401k”, which connects to internal information, but some are more like “how do I link a calendar to calendly”.

The one thing I’d love to have in the search product is a cruft cleaner for the results of web queries. Where you have cached the data presumably this wouldn’t add much overhead. Reduces what you have to feed to the LLM downstream and might improve the embeddings performance.

Insane as in “how can Meta be so crazy as to commit themselves to AR still just because it’s in their company name” or as in “insanely good”?

Genuine question!

This was one of my favorite features of Google Inbox (RIP). Does anyone know of a good browser extension that can reproduce this feature for GMail? Preferably one that just works with localstorage rather than forcing you to put your notes in some third-party service...

+1. I had simply never considered that one reason to not be a user of the dominant technology (writing, agriculture) was that adopting would mean losing control over one’s group’s destiny.

Gives you a different lens on enterprise software adoption and implementation!

On mine you definitely can make it so that there is no discernable backlight even in pitch darkness. Whether somehow it's still consuming energy I don't know. However I agree that generally the Oasis battery life leaves something to be desired.

It seems like this is all compatible with the root cause being parental/family values of achievement rather than the school per se.

As a parent, I’m always super conscious that one is tempted to praise the achievement, but this is good summary of why that is a bad idea. Instead we try hard to praise the effort, especially when our kids fail. Not always easy!

If you read the OP you will see that in one example that’s exactly what they do do - in order to reclaim the space occupied by HVAC and turn into more apartments!

+1

I'd quibble with the title. Based on my experience in London, most middle-class/white collar workers love car-free living. Most blue-collar workers don't like it. They continue not liking it once it becomes reality. (Although maybe they realize that it's not as bad as they feared).

I think this is basically another example of political polarization, combined with employment patterns. White collar workers broadly speaking place a higher value on their time and mostly belong to "tribes" that value their health and environment. Removing cars is great for that as bikes give control over journey times (and in most cases in e.g. London reduce them) while positively impacting the other.

Blue collar workers may not be able to do their jobs with just a messenger bag containing a laptop and airpods. Also, their political identity has been dragged further towards the radical-individualism end of the spectrum where collective action on e.g. environment is hard, and trust for bossy health authorities telling them to exercise more is low.

The slightly strange side effect is that you end up with an item which is way cheaper (a bike at e.g. school dropoff vs a car) that becomes a signifier of social status. I am sure there are loads of examples of this phenomenon but none spring immediately to mind!

I've been thinking about doing something a little similar in spirit to this, namely smart payment credentials for LLMs to protect against misuse in situations where you have an LLM that makes buy/no buy decisions.

The idea being to make sure that a payment credential has been requested by a legitimate chain and only then provide a single-use token (or similar).

Is there anyone working on agents that can consummate transactions out there who is thinking about this area that might like to chat? Email address is in my profile if so.

It's such a pain keeping track of all the `defaults write xxx` stuff when Apple updates.

My current setup here in case useful to anyone: https://gist.github.com/dbuxton/3ccdfd88b05e1f2807c05fe32f50...

(If anyone knows how to show bluetooth in top menu all the time on Sonoma I'd love to learn - haven't done much investigation but option seems to have disappeared in the UI...)

Agree with this. I would say that the rate of progress from Mistral is very encouraging though in terms of having multiple plausible contenders for the crown.

I don’t agree. RAG works just fine for this use case if you have a decent chunking methodology, good prompt design and some secondary safety rails. We do just this in our product (https://hrharriet.com).

The challenge Microsoft have set themselves however is that their problem domain is very large. They have to accommodate many different types of queries. We do benefit from having a specific type of usage pattern and much smaller data than a “real” Onedrive.

That said I am baffled that this product has performed as badly as it did in the review. These seem like pre-alpha type issues. I guess we should congratulate MS for being willing to put these things out to test at early stage but… come on!