sandbags

As a datapoint: I have routinely been using Claude Code through a VPN for 9 months and am in the UK. I think it has to be more than this. I sometimes shift endpoint mid session and worry I might get flagged. It hasn’t happened yet at least.

I can almost hear the “famous last words”…

The AI labs have somewhat the same problem that publishers have. Essentially their asset is a static piece of IP: a huge file full of numbers. They’re like a publisher only letting you read their book through some scuzzy Flash reader UI because they have to protect that file at all costs. At some point the weights get out/get reproduced and then what they have a is bunch of sunk costs.

I've never really used spaces, I remember them being introduced but they didn't really take for me.

Anyway to see what all the fuss is about I've installed Gridlion but… how can I move an app to another space?

I was expecting to be able use to a keystroke to "flick" an app up/down/left/right onto another space.

What would be required for you to see it as a policy?

When you use the word “allowed” it becomes a policy.

I bow to you 7 times

In fact now I think about it my claim to fame used to be that Github used one of my Rails plugins. I had written a really simple versioning system (Rails 2 I think) that I used for my blog and they used it, IIRC, for versioning wiki pages.

Surprised to find I am #79.

I think that was down to being in a particular IRC channel when CW & co. were building it.

No VORP calculations in US tech I see.

My Dad and I are doing a generative art experiment with his HP-7475A pen plotter.

I wrote an application called Axo which is a graph-based programming tool that outputs paths to be converted into HP-GL, SVG, or G-Code (we've used this for engraving with a laser cutter). Axo is an homage to Reaktor, Patternodes, ArtMatic & Max, all applications I have enjoyed using.

Here's a gallery of some of the bits I am happiest with so far: https://axo.mattmower.com/projects/mausart/gallery

At some point I want to pair the virtual image with a picture of the real-life plot. Some of the plots look great on nice paper.

It’s a problem that the models are moving faster than they can be usefully tested. GPT-4.1/4o/o-1 are not SOTA for some time and they don’t even seem to have the Anthropic models as part of their study.

Even if their conclusions were valid at the time they did the work, it says frustratingly little about it today.

We’re testing implementations, not principles.

But we’ve been here before. The agile movement originated as a response to the multifarious problems of big design up front.

Pretty much. Xcode was quirky but it still is. But the frameworks were well documented and 1 Cocoa book could get you a long way. I loved building Obj-C/Cocoa apps back then.

My understanding that attempts to defeat fingerprinting are often useless because they can tend to make you more, rather than less, unique.

So instead I wonder if we could build an open database of “identities” that our browsers could clone.

That is your browser deliberately reports the whatever is currently the most popular of a set of general identities.

After implementing a behaviour tree library and realising the power of select & sequence I found myself wondering why they aren’t used more widely.

I’ve never done anything in crypto but watched in horror as people created immutable contracts with essentially Javascript programs. Surely it would be much easier to reason about/verify scripts written as a behaviour tree with a library of queries and actions. Even being able to limit the scope of modifications would be a win.

Okay that makes a kind of sense in the case stated but even there seems open to creative abuse in cases where those lines are not wholly independent. Thanks for the information.

Modulo I haven’t tried it yet it’s been an irritant that SIP broke tracing so this is a welcome development, thank you.

I'm not in the US so I've no dog in this race only curiousity.

I can understand allowing a governor to change the text of a bill. But I cannot understand allowing them to sign those changes into law. It seems like that would mean they could creatively reverse the meaning of any bill.

It seems like a governor should be able to approve the text as written, or change it and send it back.

What am I missing?

I'd forgotten all about it but SubEthaEdit was such an amazing tech when we were using to collaborate internationally back in about '04. It went off my radar but I am glad to see its still available as a free app.

I'd argue that privatising telecomms has been a win. I think because it's quite open to competition (even if there has been a good deal of consolidation).

The challenge with so many UK privatisations is that the idea of real alternatives/competition ranges from laughable to extremely laughable.

I don't think that's what he's saying precisely.

I think what he is saying that native platform apps get delegated to different teams and coordinating among those teams becomes an additional cost. You don't want each team going off and doing their own thing.

Your 'answer' is "use a cross-platform GUI toolkit" but that has its own challenges. Not least that you typically build a native app because it delivers a native experience that users expect.

In general (and I accept there may be counter-examples) cross-platform tools fail to do this.

I don’t have a dog in this race but I was also around at that time and one reason is there was far less choice in 1995 about where you would go from C. C++ was also a vastly simpler language back then (no templates, no exceptions, barely a few hundred command line options). So I am not sure what its adoption then can teach us about language adoption now.

and (and as much as I do love Alan Rickman) more properly Stephen Moore.

You’re right. That aspect of how Brexit was carried through was not acting in bad faith. The anti-European faction has been fighting since we joined to reverse it. Many other aspects of the process were in bad faith but people must be allowed to change their minds, disagree, pursue their faith.

I would love to replace JS as the scripting language in one of my native macOS apps and I wonder if Wren would be suitable.

The two biggest questions I’d have are:

1) how easy it would be to bridge Obj-C objects to Wren-space and vice versa (a big win of using JavascriptCore) 2) how easy would it be to implement script debugging? This is not exactly a strength of Javascript core but it is at least possible by connecting the Safari web inspector.

There’s lots I don’t like about JS and JSCore but i’ve yet to find a better alternative.

Likewise. It still works fine for all the things I use it for. Long may it last!

I want to say “prediction is hard, especially about the future” and it is no surprise that the predictions of a model built in the ‘70s might not hold some 50 years later. That they held so well for so long should, perhaps, be the headline.

Does Forrester’s model get updated and re-run?

It's a fair point that nationalised water industries can also be poorly run. But I'm not sure what the argument is that means the amount of money that privatised UK water companies have paid in dividends vs. invested in maintaining and expanding infrastructure isn't a significant part of the UK's problems.

However, as a further point. If national priorities change then a nationalised water industry can respond (relatively) quickly. But what can be done with a bunch of potentially foreign owned profit-seeking companies?

As the article mentions, privatised water companies have built no new reservoir capacity and relied on drawing from rivers and other sources.

What the article doesn’t mention is that pre-privatisation a new reservoir was built every year up to about 1960 and then every few years until privatisation in 1992.

So we are about 30 years behind in adding capacity to the system. This combined with the inadequate levels of investment in the system leading to enormous wastage, is the answer.

Water should never have been privatised. At least not without a framework for a national strategy for water. I suspect that wasn’t done because it would have made water companies and unattractive source of profit.

I saw someone reference this today and the question I had was whether this counted the trillions of words accrued from books and other sources. i.e. is it 40%? Or 40% of what they can find a direct attribution link for?