jgord

11 ms·

121.
by jgord 2y ago
At first glance, it seems like _porting_ code to another language / api .. is probably a good sweet spot for these code LLMs .. .. and even automating the testing to check results match, coming up with edge cases etc. ... and if thats
122.
by jgord 2y ago
A container-style standardization of apply-for-capital process, might help make an open market for small-startups meet small-investors work : standardized questionnaire / pitch format standardized terms trust ratings on both
123.
by jgord 2y ago
Ask all the Qns and have a fast fits-reality filter ? Seems like a general recipe useful in startups, science and life. afterthought : We are not doing this now, either in science research or in startups : The current VC model of funding st
124.
by jgord 2y ago
what %ge of C++ features did you use for that ?
125.
by jgord 2y ago
Glad to be proven wrong in my subconscious assumptions about authors age and sex .. I assumed some bearded 50yo polyglot who came up in the 80s writing TSRs in A86. About page does not disappoint.
126.
by jgord 2y ago
The first section on rust protections static vs dynamic is thought provoking and informative... tbf, author admits he needed to dive into ASM to work around the C compilers sensible optimisation defaults. tbf to the author, he does explain
127.
by jgord 2y ago
awe inspiring photos .. and we really need awe-inspiring right now.
128.
by jgord 2y ago
Are we so beaten down that we think no new useful technology is possible ? That we cant do large projects ? It is quite hard to not be cynical .. we have lots of evidence that we should be cynical, that our politicians are scamming us etc.
129.
by jgord 2y ago
Thomas Piketty lays it all out in "Capital in the 21st Century" Basically he compares two kinds of growth rates : the growth rate of the 'real' economy and the growth rate of wealth itself. You need the wealth rate to be
130.
by jgord 2y ago
The history of science is full of false starts ... Ivermectin may or may not be effective against various cancers .. but it is a very safe molecule. Look at the research independent of popular and institutional opinion. If you can be part o
131.
by jgord 2y ago
Bruce Schneiers thoughts : https://www.schneier.com/blog/archives/2025/02/doge-as-a-nat... https://www.schneier.com/blog/archives/2025/02/ai-and-civil-...
132.
by jgord 2y ago
I did lookup the BTC transaction fee, but it looks like 15 USD was completely wrong .. my bad. Checking now : https://ycharts.com/indicators/bitcoin_average_transaction_f... This says, afaict, the average BTC transacti
133.
by jgord 2y ago
Many of us in the BC is digital money camp argued the case that more transactions and a lower transaction fee could be most easily implemented, within the spirit of the founders intent, if we simply increased the size of the transaction buf
134.
by jgord 2y ago
At the heart of all progress is the mantra that "best idea wins". Maybe DeepSeeks creative use of RL within LLMs will open up founder and VC interest in using RL to solve real problems - I expect to see a cambrian explosion of hig
135.
by jgord 2y ago
I keep telling VCs and Angel investors that now is the time to look for relatively under-valued high growth startups using Reinforcement Learning to solve real world B2B / engineering / logistics problems ... The balanced NPU&#x2F
136.
by jgord 2y ago
My opinionated view, based on the ancient katra of code practitioners - seek smaller code, fewer dependencies : When moving from PHP, better to ditch 'modern' over-engineered compiled javascript, and instead use javascript at firs
137.
by jgord 2y ago
well thats perfectly logical IF you want to prevent under-16s from using social media.
138.
by jgord 2y ago
I propose 'gords rule' : "any sufficiently advanced LLM will learn the laws of logic, the principles of scientific method, and Reinforcement Learning" until that happens .. I think RL startups focused on real problems ar
139.
by jgord 2y ago
I definitely view the browser as an app delivery system... one of the benefits being you don't have to install and thus largely avoid dependency hell. Recently I wrote an .e57 file uploader for quato.xyz - choose a local file, parse it
140.
by jgord 2y ago
readable .. a useful variant on "aim to practice in your zone of discomfort to maximize learning".
141.
by jgord 2y ago
We will soon face another kind of bit-rot : where so much text is generated by LLMs that it pollutes the human natural language corpus available for training, on the web. Maybe we actually need to preserve all the old movies / document
142.
by jgord 2y ago
scary good demos.
143.
by jgord 2y ago
CSV is kinda great .. but it does help to have nice tools to wrangle it, such as the famous xsv by burnt-sushi.
144.
by jgord 2y ago
btw, thank you for this beautiful project.
145.
by jgord 2y ago
He is a much beloved gentleman rogue. Invariably well-informed and well-spoken, even if somewhat self-centered or arrogant at times. For him and his family, Im glad hes free. Five years seems a pretty harsh sentence for publishing leaked in
146.
by jgord 2y ago
Would it not be more efficient to put solar on every bunnings/warehouse/mall rooftop and over parking lots .. rather than on home rooftops ?
147.
by jgord 2y ago
Bring the risk forward .. if there is an easy to identify " most risky/gnarly " part.. aim to get a proof of concept of that done first.
148.
by jgord 2y ago
Looked up surfels .. I think they are different from fitting local planes - I was interested in the geometry, not the lighting. But surfels seems like a clever idea, pre-dating view dependent NeRFs.
149.
by jgord 2y ago
Im quite familiar with potree and a big fan .. having hacked some of the internals and added features to my own custom version - so people can share annotations, measurements, save to cloud or export linework .. without writing code/cu
150.
by jgord 2y ago
ahh.. looking for investment, so reticent to talk about the internals. Basically does a lot of matmulls and sampling to find good good planar fits... Most of these sample datasets took an hour to compute, using 12 threads on a midrange CPU.

More ›