nxdmum

Qwen is not available in India . been wanting to try out that model. anyone from Qwen listening ?

It's about the complexity of a single cell - the number of mechanisms it has in place on how it works. He sets up the argument for Intelligent Design using the Cell.

Anyone read Signature in the Cell by Dr Stephen Meyer

IF you havent written your own harness - you would not understand what you can do when you're writing your own harness . the current set of harnesses - all of them are crap tier. The only clue i can give you - it's not in the model providers interest to have token efficiency - but when you are coding the harness yourself you can shoot for that .

In today's world - and idea stated stated is an idea stolen .

Mainly because the US Dollar debt at some 39 to 40 trillion is now unserviceable and the US is going to default the debt payment. So countries are preparing for that eventuality - the US ship is on the event horizon but being slowly but surely sucked into the Black hole. The last govt was ensuring spending all the unspent budget - even if they knew it was bad for the country . The current govt will be left holding the baby when the crash comes. The only question is how long can you hold - 1 year 2 years before the ship crosses the edge of the event horizon

If you're on their subscription plan - agy cli or antigravity ui is the only choice i think.

Anyway - if you're a dev - you would be writing your own agentic env right ? that's the best way forward. I wont tell you more than this . but if you're not - you are losing out .

so far using Gemini from 2.5 Pro to date (3.7 flash) - the way google trains the model it seems - is to identify top 3 to 4 things to fix first. as a result gemini is not that good in being thorough - but it's a needle mover . Opus5 Opus4.8 and Fable always point out things that Gemini missed. but Gemini was a needle mover - identifying the most important things to fix.

I always enjoy interacting with Gemini . When i ask it questions about designing a new model etc - it's always the most helpful and encouraging . I really want to thank the Google team for this and their happy positive models they generate.

I use gemini flash after a round of deliberation between Sol and Kimi these days on the main plan . Kimi 2.7 paired with Gemini 3.5/4.6/3.7 flash has been my implementer - Kimi k3 and Sol 5.6 have been my planners and code reviewers.

I would have ideally like Anthropic and was on their USD 200 plan - but after they didnt sign the letter for Open source models - i dropped my subscription . Wont make a difference to their lives.

But Sol is great . Combined with Kimi K3 for adversarial plan reviews - you get robust plans . And Sol as a reviewer for Gemini/kimi 2.7 - you get great edge case handling and robust code.

I also integrated muse 1.2 - on the contributor tier - and I Just saw facebook release 1.3 muse . this is great news. The training im permitting is my thanks to FB for releasing the open source models of the past ! Thank you !

Simple question - can i use Gemini 3.7 flash with a subscription in my own harness and not in agy client ? You're from Google so the question.

By the way - I love Gemini's personality . it's phenomenal to work with

The reason i want my own harness - is the custom tools that i provide vs the low tier tools that come with the custom harnesses.

You'll would really benefit , if we could use Gemini in our own harness and not be forced to use it via agy . i've tried using gemini to circumvent - but not been successful.

If you see this - please reply here

Thank you Hillel Wayne. I did the your TLA+ book twice . I am sure this will be a great book.

Google Gemini agy is not allowing you to use your token via your own harness. My own harness is far more efficient than agy. They can take a simple stance - if you exceed your token limit they block you - with the 5 hr limit they are already doing this . so there should be no reason to block you from using your own harness - if you are more efficient - you gain - if you are less you lose .

They are not allowing me to hit their endpoints which agy hits - it's frustrating . i tried to hack it with gemini itself. what i love about gemini is it's so encouraging and ready to help you - even against the agy client : ) .

Even though im so frustrated with this - i still love Gemini for some reason ! Most encouraging model in the world!

God Level. 1 level below Ling Ling

Yes . I was showing a friend the prodigy video and the next day I was reviewing their series on the Hardest violin pieces - Easy , Medium, Hard, God Level and then they had Ling Ling. I was watching it for the Bach Chaconne - I knew it was marked as difficult but didnt know where. And suddenly it went private while I was in the midst of watching it.

PS. I had just watched James Ehnes Chaconne on youtube . It was beautiful https://www.youtube.com/watch?v=xjYQlmpS69k

How do people who do things like this - get the time to do it? Did he do it in his personal time with or without a job? If anyone else does something like this - can you give your insights. I would love to be able to do something like this.

What PCB design software did you use for the PCBs? KiCAD or did you go for something paid from Altium.

NoThisIsJohn typing at 200+ words per minute:

https://youtu.be/oOdfefV2R1I

Yes. I would switch over when i reached 70 words per minute.

The initial mental effort is high. But the trigrams are hashed to the letters on the keyboard. so "con" is hashed to for example Lower + C. So that helps. Also - after sometime you start seeing the common 3/4-grams - tion , the, and etc. Then it starts to become like a game with a dopamine feedback loop. Every time you hit 2 keys and get 3 or 4 letters out - you start getting a high.

>Out of curiosity, what is your motivation behind the experiment? I watched a video of NoThisIsJohn typing at 200+ words per minute. I know i cant hit those speeds, but i was wondering what hacks i could do to the keyboard to get as fast as i can.

I just wanted to hack typing. I even thought of double pressing keys - for example bigrams. I came up with a new layout. When i read about the norman layout - it is criticised for same finger bigrams. But what i thought is, if I put the bigram keys near each other - then I would be able to press those with a single finger press - if the finger overlaps the 2 keys on the edge. but for that you need key wells - even at the edges of the keys - to give that tactile feeling. Similar to what the thinkpad has at the edge of the G, H and B keys . This would allow you to press 2 keys (and maybe more with a correctly designed keycap set) at the same time. But I would have to design new keycaps for something like this. And I thought of writing to Signature plastic - but I dont think i have that much money right now - but maybe a little later.

I also thought of 3 dimensional approach. Which is to insert vowel keys between 2 keys - exactly like the thinkpad Red trackball - but at a lower height as compared to the bigram consonants. This would require a special MCU and n-key rollover detection - to check if the 3rd key is pressed. If you want to press 3 keys - you would have to press deeper - so that the deeper sandwiched vowel key also gets depressed.

It would need a new PCB design and I started reading up and watching videos on PCB design and Keyboard design.

Anyways, If any Keyboard designer is reading this - reach out to me

I was consistently hitting 40-50 on MonkeyType - but while transporting the keyboard - i damaged it a bit. so now out of touch. I was practising about 20 mins every day for about 50 days - with misses on days when i had a lot of work. But i know that my speed would increase. Now when i see words - i can see the trigrams and I can visually break them down.

I am confident i would be able to reach 70 if i practice more. The main issue is that my speed (40-50) feels so slow that in my day to day - I used Qwerty and had not fully switched over.

I'll state my requirements - for anyone interested in meeting them:

I have an OLKB preonic. And I used QMK to significantly modify my layout. I started out with the modified Colemak - but the re-arranged the keys to move keys which require the index finger to move to the side - on the QWERTY that would for example be G and H . I moved B and K forexample where Z and X are on the Qwerty - and moved Z and X to the number row. The numbers itself appear as a layer when i need them.

Peter Norvig has an article called mayzner : https://norvig.com/mayzner.html - i wrote a program to analyse the frequency of bigrams/trigrams, 4-grams and then programmed them as shortcuts to the keys - using the lower / raise

This means i can type conversation as : con ver sa tion - where con is a trigram , ver is a trigram and tion is a 4-gram. s and a need to be typed out.

my needs : 1. MCU should have more ROM for writing the new preonic has 64KB. 2. Columnar layout - Split keyboard would be ideal.

Reasons for more ROM - you can program a lot more common words in it and Columnar layout would be easier on the fingers.

I've checked out a lot of keyboards - The Iris is good - but the ROM is only 32Kb. The Moonlander is too expensive for me - i cant justify the price to myself right now.