deskamess

> Nvidia being down 18%

The only part of DeepSeek-R1 I do not like. I hope it's over, but I am not holding my breath.

What CPU and memory specs do you have?

Doing its own PR's as well. This model is now marketing itself.

Connections is a great game to test AI. It really relies on the ambiguity and loosely connected aspects of culture and language. I am shocked at how well o1-pro does.

I think they can work well if you do a lot of city driving (where you brake a lot). So in a sense, it's the best of both worlds.

They are cheaper than pure EV's and do not require the home charging infrastructure.

Let's face it, the Canadian reaction is purely out of US friendship. There is no end-to-end EV car manufacturing in Canada. There is an EV battery setup in Ontario (and perhaps Quebec?) but that's about it. We are a decade or more away from having an end-to-end manufacturing pipeline. So... we implement tariffs that hurt the majority of the population? For the govt's lack of investment across decades. If you want to, put a tariff on EV batteries or any other part that is manufactured in Canada - not the whole car. There is no good reason for us to take this tariff.

At the same time, politicians will talk about the dangers of climate change and how we should all try hard to mitigate it. I guess some solutions just don't have the right story to it.

I do agree that the situation we are in now is due to lack of investment in EV/associated tech starting 1.5-2 decades ago. China has invested over 200 billion in EV+Solar and are reaping the rewards.

Would this be the case even if you were moving toward or along-side the 'reflector' (black hole/other body)? For the sake of discussion assume we are at or beyond the focal point.

Complete newbie on the topic of GL but I find it very interesting.

Q: We seem to want to use our sun for GL experiments. And that requires sending something out 500 AU out for focal length purposes. My question is somewhat theoretical so assume there are no alignment issues for the following thought. Are we not already at the "focal point/line" of some other 'entities' (sun/heavy body/galaxy/etc) gravitational lens?

Hmm... is that what this paper is talking about - the fact that the GL observation was made from Earth on some infinite focal length. Listening to other YT videos on the topic seemed to imply the resolution would be much higher. Or is the resolution location/technology related?

> the ability to scribble notes, turn back the page to re-read something to double-check without much effort.

I agree 100%. The simple act of turning back the page/pages is so fast and efficient. My brain seems to even remember where on the page the info I want is. So much more is visible in a single 'eye-shot'. I don't get that same mental experience with e-docs - not sure why. I did grow up with paper so there is a bias.

How about BYD EV's?

Is ESS an OpenID compatible server, or does it integrate with OpenID backend? Looking for an OpenID provider for a small-medium friends & family group, so was hoping this fits the bill. There is a dearth of information on this aspect of the product line.

The first person (EV1) was awesome as the Earth was visible (day-side). And it slowly went night-side. I was surprised to not see any lights at that point and that was possibly because it was over the Pacific - or the actual vehicle was pointed away from Earth.

That image of a human profile with earth in the background was kinda nice. The 40-50s lead up to the profile image: https://youtu.be/ABQBEdOzrV0?t=6868 . And as others have pointed out, the suit looks rigid.

Is there a pre-intro version of this document? I am lost at the Introduction itself and not able to get past it. In particular:

> If I need to position child items within the grid, I need to specify the line number for each item by using grid-column.

Stuck at what is 'line number' in this context? I tried to force my way through it hoping context would fill it in but immediately got stuck at fractions like '1 / 3' followed by '1 / 4'.

Gov of Canada RTO is creeping up just fine. Now at 3 days for some groups.

Got to support your restaurant businesses and parking lot donors. Never mind the insufficient capacity at the offices when it was just 2 days (they sold a bunch of buildings first!). Never mind the poor public transport in the capital (esp after LRT; instead of 1 bus, its bus-train-bus) - and it performs really well in the winter - its gonna be great. Never mind the people who now have to drive in traffic inconveniencing themselves and those who actually have to be on the road. Never mind the eco-footprint of their decisions - given how much they cry about how 'Canada should do more to fight climate change' (so virtue signalling at its best). Cant wait for winter when overcrowded offices causes illness that we can take back to our spouses, kids, and grandparents - I guess this should cause some natural attrition (the deadly kind).

No surprise that I am a very strong proponent of messy desk. Organized chaos.

As one of the commenters to the video said:

"That is the Noon Flight from Moscow arriving as usual!"

Nothing to see here. Some bounce cars - others dutch roll planes. Personally, I am partial to the Swiss roll.

Build your own web form: https://agnikul.in/#/

Select your launch location, orbital inclination, payload mass and then 'Build'.

And another more detailed form at: https://agnikul.in/#/book

History Bites was another good one.

Will that 90% model give you a more accurate answer after 3 tries?

Looks good but 10 kg/23 lbs? I see a Samsung IPS for <8 kg. 50" 4K Fire TV for 9.3 kg. I guess we don't really move monitors around once we start using them, esp if mounted, so perhaps a non-issue.

Their parental control system is not too good either. For a box with multiple users (some kids, some adults), its almost all or nothing in terms of game access. Tough to sub-list when the main page is all the games unfiltered. If they could somehow fix that it would be good.

The RAG explanation was pretty good. It had the right level of description and it explored the sub-topics and defined them.

Based on the TOC it seems pretty deep. At this point I am looking for something high level.

As far as definitions go that was actually pretty good.

Side track: There is so much going on this space. I wish there was a chronological flow of a machine learning scenario/story with all the terms being introduced as we meet them (data, pre-training, training, inference, mixture of experts, RAG). Like someone walking me through a factory explaining what happens at each stage (like Mr Rogers used to do). Most of the time I do not know where the terms fit in the big picture. When I first came across pre-training I thought it was something done to the data before training happened but it was actually another training.

Do you have other eye-wear for the day? My daughter had started with regular glasses for near-sightedness which we got quickly when we realized she had trouble seeing in school. This year we did a checkup and the optician recommended specialist glasses (MiyoSamrt) as my daughter had no experience with contacts. I am now wondering if she should try something like RGP at night, and then use her glasses for short-sightedness. I know some treatments can be additive and some other combinations are not recommended (like drops + glasses).

Hijacking this thread to ask a question. Does anyone have a 1000 ft level description of all the steps related to the ML pipeline. I thought I had a basic understanding, but I am now seeing things like pre-training which I am not sure about (why pre-train, and what are you pre-training with). And fine-tuning which I understand better (at least the concept). I guess I am looking for a concept level definition of the verbiage/taxonomy used in the ML pipeline - ideally in chronological (pipeline) order.

> Desktop Destroying Screensaves.

There is a guy on YT who does these mini-anime (multi episode) of stick figures fighting each other. The opening sequence happens on a PC desktop against a virus. It is just amazing and that whole sequence as an animated gif would make a great screen saver. Now if it could be made dynamic so it works on the existing screen (like xmelt) it would be great.

I would actually settle for the original (non dynamic) one. The one thing you would miss is the sound effects which are also great.

I regret getting rid of it while moving to a new house. Even though the screen had turned green-ish it still worked. Had the newer Wiimotes, balance board and all that. Miss it terribly. The Switch is good but Switch Sports is not the same. No planes to fly, no mountains, Showdown, etc.

They have been serving those on airlines for a while now. A solid choice. Hopefully a 'BuyItForLife' kind of choice for the airlines.

Wonderful on planes and wonderful at home. Also great when dunked in milk. Make sure you let a small bit (or one whole) 'accidentally' fall off into the milk and rescue it later. Hmmm....