davidy123

I don't use Vite though. Aside from typescript, I'm trying to be as vanilla/unopinionated as possible, using Web Components for example. And there are hundreds of Jest tests I don't want to port.

I'm going to keep hitting reload on this page until someone solves some of the worst problems of my work life right now; getting a npm monorepo with typescript, eslint and jest working smoothly. Thank you to whoever(s) that is. I'm trying to make a public project, but building/developing with it is successive layers of sometimes-works voodoo that have nothing to do with the project goals. Somewhat compounded when members of nx (nee lerna) popped in to try to help, adding another layer of sometimes-works. As it is, DX depends on a fast computer to basically rebuild the whole thing each time, or running build & test in a dozen workspaces, because I haven't found any other reliable "workflow."

I tried to use Bun but it doesn't support some low level Node APIs properly, in my case for Playwright. Keeping an eye on it though, what I saw seeemed promising.

The problem is they are chasing a 1990s vision of computing, office suites in particular. They're just creating more silos. By this point, they should be working with data and schemas, not documents. Documents are just a by product. Open Source has an embarrassing tendency to think this 1990s vision of a desktop & office suite is an end point, but it's just a distraction Microsoft will always dominate. Open Source would have been better off trying to create a futuristic 3270 but more seriously taking the Web design to heart (including Linked Data).

People are just not that good at organizing in groups and information is too diverse to leave to individuals to categorize in blobby programs with arbitrary fields in hierarchical files and folders, so inconsistency and disconnected work is the result. This is at the height of its bad in government. Now we are at the point where AI can serve as a connector, with careful use (it would have been much better if organizing data had been the main priority of people). For example, referenced documents created by AI on data with a final group review, the perfect use case of efficient and effective AI integration. But Microsoft is at the top of that heap, in terms of information capture (even if it's not accessed by them directly), capability and authority. Meanwhile Open Source is still talking about a new desktop music player and how to use grep.

Apple Silicon made x86 silicon look bad, but what has it really upended? Macs are taking over more of the personal computer market, but hard to say what the factor is there. I think it's mostly network effects, partially due to their shameless proprietary approach. PCs, Apple or other, are kinda generally good, no matter what the price or combo, and disappearing at the same time, a lot can be done with just a browser on any foundation. Apple seems to be years behind or nonexistent where things are really changing, AI and cloud.

In my experience, it leads to slow, expensive solutions. It might be really helpful if it provided special purpose small, fast handler models that could be run locally so openai would only be used when really called for.

I don't know who this mysterious person who would want to see my photos in 50 years is, but I keep photos for myself.

Storage is cheap enough that I really don't worry about paying for it, and I don't think that's going to change at my scale.

The privacy and control issues are something to worry about, though.

I don't understand why more people don't care about round-tripping. Yes it was over ambitious and far too heavyweight in the 90s, but today you should be able to generate OpenAPI and its data and component equivalents from the visual graph, or inversely import your graph, then verify / update either "side" at any time. I've mentioned this before, but TogetherJ was the best example I saw, lightweight and worked well for visual/code round tripping when it was all in a predictable Java environment. Somehow we went from overdone to decades of "works for me" mucky muck.

Thank you for this, but I guess it's out of date, for Canada at least it's now https://articles.alpha.canada.ca/forms-formulaires/

It went from DIY open source to something more like the UK's model; in fact Canada is running with a number of UK-origin projects & ideas, like Notify, which is of course a good thing, we should all learn from each other's best practices and innovations.

That might be part of their calculation. After five years or so, most people aren't going to want them. So they might run poorly, but at least they're not abandoning them. Anyway, it was mostly Qualcomm that was preventing longer upgrades, so I respect them for finding a way out of that. Maybe Qualcomm will now have to support their chips longer, due to new expectations.

I like the change from "master" to "main" and a lot of other changes of our times. And I have found it very often says something about a person when they dig in their heels and make a lot of noise about it. Many words are too often used trivially and inappropriately, often when they're not even technically correct (the relationship is very often better described as "controller" and "worker," for example).

This is a moment when one of the big projects is to recognize and address bad habits that have been carried forward. And I especially like the opportunity to find more appropriate and considerate words. Yes, it is sometimes merely a nod. If that is a conceptual monument of our times, that makes some texts look inconsiderate and dated, I'm good with it.

I don't know why they don't mention the excellent WebPilot plugin, which has been able to search the web (on any browser natch) for a while. It has all kinds of nice open source abilities.

Yes it is useful as a deflection against early hype, but I don't think the headline is fair.

I've learned the Register isn't a very good rag, they push the cattiness so much, you can only get information from between the lines. The reality is "AI" is an important part of the future, local AI if it's going to be a good future. The other reality is Apple is way ahead with its M chips for consumer devices, with their tremendous memory bandwidth and GPU well suited to running LLMs up to the speed of a 4080. If they didn't limit RAM so much and could scale up to the level of professional GPUs, they would be very compelling.

AMD is being very flaky with their rollout. They only provide the ML chip on some of their new chips, and let some vendors block it out in BIOS. I've been watching their github demo site, and they won't commit to anything but a very limited demo for Windows.

Intel at least released an open source kit earlier this year and will support their VPU on all Meteor Lake+ chips, as well as other processors. Consistency and openness is needed to break NVidia's deadlock here, OpenVINO could be a good step.

Meteor Lake will have 128GB/s for on-chip RAM bandwidth, that's a fraction of M-chips, it really remains to be seen how this will shake out, but it seems likely there will be a new class of computing device, and it may make all others very second class.

This reminded me of the syndrome of descriptions of software that assume knowledge of the last system. eg "well-known Linux facilities." Pretty sure the number of people who know what network namespaces are is smaller than those fuzzily familiar with container networking.

I'm not sure about unnerving. Psychologically, I know I'm looking at a recording. But when they're not looking at the viewer, it looks like they've got better things to do. I see this being used all the time recently, including on large productions. I assumed it's meant to convey the feeling of talking with someone who may also be talking to others. But it gives me hives (for which I'm losing points, but what can I say).

Nearly totally unrelated, but what is with the recent trend of showing people talking to but not look at the camera? Do they save money that way? It drives me bananas and is just a distraction from the content being conveyed.. Altogether, with their obvious attempt at slickness, I find this video rather corny.

The number of people who disable Javascript is probably less than people who refuse to or can't upgrade from really old browsers, where all the current HTML and CSS is going to break.

I don't get the HN hysteria over Javascript. You have a powerful host (I won't even call it a client, because it can and should be so much more). It would be crazy not to use it. Sure, it can be abused or programmed poorly, code should be tested. But in my wide-ranging use of the Web which involves a ton of Javascript, that has rarely been a problem.

When are we going to get away from all the YAGNI engineering that has sold short the promise of the Web for more rented silos, and get to something more like HATEOAS?

I find this video discussion of Orwell and his works to be illuminating: https://www.youtube.com/watch?v=2Gz0I_X_nfo

Asimov's critiques of his work, for example.

Do you mean you don't use NVLink or just use one that works? I am under the impression it is being phased out ("PCIe 5 is fast enough") and some kits don't use it.

I didn't mean "but" we need that, I meant "and" we need that. In case there's any doubt, I am 100% for immigration.

People are living longer and healthier and more independantly, which means they want to stay in their city homes in their 60s - 80s. What do you want to do, put them in warehouses? Plus there is a lot of immigration, but we need that.

There is an ongoing process of densification, it can be faster but it can't be done too quickly, it's not fair to blot out people's views or destroy neighbourhoods or destroy every remnant of past eras.

With these dynamics it's no surprise that city housing is so expensive, and I don't think it comes down to one factor like interest rates or greedy people, so that billboard is just political pandering. What would your attention-getting measure ask for?

I get the impression a lot of people are complaining they can't afford a family home basically in a prime-ish city location. Can't blame them for wanting that, but an option could be to move outside the cities. Properties are quite affordable, and there are some good transit corridors and decent towns (and advantages of living in the country), though better commuting options will always be welcome.

This is what I've done, an hour outside a city, I'm on a forested acre with a lake, with beautiful views year-round. I'm far from house poor so I can afford to travel all winter or do overnights in any city.

To enable this, I would push for better commuting, with more hubs and extended service so it's super easy to get to downtowns. But resolving scarcity of talent in the trades would also be key.

Huh! I hadn't seen dataview before (I think mdld predates it), I'm more than happy to use another work if it's a supported superset. mdld is sort of inspired by Semantic Mediawiki, dataview uses SMW's :: notation, which is a good sign. However, I don't see a few features that mdld has, like concepts, which are very useful, and the hedgedoc application of mdld has some nice dynamic views, for example, creating a gantt chart out of data dispersed across markdown files. In general, its emphasis is creating things out of data to support features like round-tripping, rather than creating self-contained black box information views. But its more of a proof of concept, so I'm going to look at adopting dataview.

It (or maybe a new version of it) is the future. But people got bored and justiably frustrated in how clunky it is, because it can be and do everything. As often happens, people started abandoning it just as it was getting really interesting - w3c xlink, the hypertext/data document of our dreams, and we're not talking about some neato mostly-great thing someone came up in popular language, rather a set of precise standards with implementations in every language. But in its era, the current "get er done/yagni" startup culture was really kicking in as people realized it was a hassle and less immediately rewarding to think of the web as a standards based platform past using the browser as a client.

I don't think people should be making tables, they should be making data, which they put into a table via a query.

I ran with this for my own note taking system, mdld. https://github.com/vid/mdld , which works with hedgedoc.

I was trying to find out how noisy this thing is, I'm guessing it won't be pleasant to be around.

I presume that's sarcasm, considering he spends a lot of energy preparing for that event.

Related: https://www.theguardian.com/news/2022/sep/04/super-rich-prep...

Yes, based on the NeXTSTEP roots, everything should be object oriented. Think of it like visual copy/paste. Pretty much every on screen item should have a meaning and relationship to everything else.

For example, in https://www.youtube.com/watch?v=rf5o5liZxnA at 35:00 you can see how an application is loaded from the network and a live chart dragged from one app to another. That isn't a one-off gimmick, it's how the OS and that version of Objective C worked.

I bought a Macbook because I was excited by this idea, but I quickly realized it was not even surface deep. I could drag some on screen elements around, but it was inconsistent at best.

But I guess there was a conflict with Raskin's idea of making everything resemble a real world equivalent (object as in "toaster app," rather than decomposed feature of "can toast") and they wanted to let third party applications do their own thing. If the whole OS (and network) was effectively the application, what's their product? Never let profound world changing innovation get in the way of a trillion dollar company, I guess.

I think of it, in my muddle headed imagineering way, this way: https://wiki.zooid.org/wiki/20120407/Raskin_vs_Engelbart

I suppose by now we would think of it all as a graph rather than OO, just add coordinates (and the required couple decades of cultural development of how they would relate) and you've got your spatial interface.