nyrikki

45 ms·

391.
by nyrikki 2y ago
It is really the only safe way to use it IMHO. Even in most simple forms of automation, humans suffer from Automation Bias and Complacency and one of the better ways to avoid those issues is to instill a fundamental mistrust of those system
392.
by nyrikki 2y ago
FYI, while these block level methods do have a use case, parallel rsync and other file level tools are far safer and often faster with less additional load on the disk. Duplicating the OS/FS behavior hits the decidable problem and you
393.
by nyrikki 2y ago
Never made that claim, but dawn and dusk are the most risky. Mine happened on a lunch break, can't see through rock either. But you only see a fraction of the wildlife that is there. If your sole stratagy is to see them you may be sur
394.
by nyrikki 2y ago
90% of drivers consider themselves above average, and where I grew up we had pronghorns, which could pass you and cut in front of you even at 55. While reducing speed when you see deer is important, you won't see the one you hit. In my
395.
by nyrikki 2y ago
Your missing the point. When your kid runs in the road. Do you want the to avoid an accident, or do you want to optimize for them just turning enough so they don't lose their insurance. When traffic ahead of you makes an emergency stop
396.
by nyrikki 2y ago
No, the inference they have isn't reliable enough to predict anything except with broad groups. Swerving to avoid road debris or hard braking to avoid a deer as examples of what have cost commercial drivers jobs. Look at the CDL world
397.
by nyrikki 2y ago
That is more of a confirmation of what everyone knows, many CEOs have no clue what SWE is. It will take some time but it will die, just as visual programming, total quality management etc... Anyone who thinks that code generation is the pri
398.
by nyrikki 2y ago
Broadcom aspires to the Oracle model, they said as much when they bought VMware.
399.
by nyrikki 2y ago
How to tell everyone that you don't know what SWEs do... They aren't slinging pig iron into train cars. If you are in the C* suite and believe this you need to find someone to feed context up to you stat. Probably why lean, Toyota
400.
by nyrikki 2y ago
All I can offer without going into the weeds is that it is not the shape of the paths, but mearly the relative length of two paths though spacetime that matters. Proper time and proper acceleration is possibly a pathway in flat spacetime? L
401.
by nyrikki 2y ago
The twin paradox holds in a Pacman universe where there is no change in direction. The main issue with the twin paradox is that it demonstrates where our euclidian intuition fails us. There are several interpretations on how to resolve it,
402.
by nyrikki 2y ago
This is the map territory relation problem. We bring in the 'Trinity of thought' a priori and forgot about the advantages and costs. You have the principal of excluded middle, principal of identity, and the principal of non-contra
403.
by nyrikki 2y ago
You don't need access to persistent storage, especially with multi-entry executables like busybox, which are often similar to built in rootkits today. I actually just spent time last week getting rid of tftpd, telnetd, netcat etc... on
404.
by nyrikki 2y ago
So visual programming x.0? I am pretty sure PLCs with ladder logic are about the limits of the traditional visual/macro model? Word-sense disambiguation is going to be problematic with the 'don't need to learn' part abov
405.
by nyrikki 2y ago
So was Internet Explorer at one time. I view maintaining android compatibility as a part of a vendor mitigation strategy, avoiding the creep of tight coupling that can make adjusting to market shifts an intractable problem.
406.
by nyrikki 2y ago
First time I have ever heard someone say my Minimoog, OB8, Prophet and modular synths weren't synthesis. ADSR is subtractive even if you ignore the filter. The (ideal) square wave contains the odd-integer harmonic frequencies, where th
407.
by nyrikki 2y ago
RFC1035 imposes a 255 character limit per string on TXT records .
408.
by nyrikki 2y ago
That is the popular myth. https://www.lisc.org/our-resources/resource/opportunity-atla... https://www.brookings.edu/wp-content/uploads/2022/06/2022_FM...
409.
by nyrikki 2y ago
Vec4 is a vector, but yes C++ base vec is not. In some ways CS is more correct with the modern (abstract) algebra definition of a vector, vs the historical didactics arrow concept. Tensor has been abused, it is a property of bilinear maps,
410.
by nyrikki 2y ago
Homework weights on grades is going to go down. Parents income and zip codes impact educational attainment more than class rank outside of the oligarchy anyways. Opportunity and social mobility is far more restricted in the US than most peo
411.
by nyrikki 2y ago
Reminder, you win by learning the material, not by completing the homework.
412.
by nyrikki 2y ago
Rice's theorem means you can't choose to decide if a program is correct, but you have to choose an error direction and accept the epislon. The Curry–Howard–Lambek correspondence is possibly a good tool to think about it. The reaso
413.
by nyrikki 2y ago
Not really related to the above post, but one thing I am not seeing on an initial pass is the advancement of understanding of problems like riddled or wada basins. Especially with time delays this and 3+ attractors this can be problematic.
414.
by nyrikki 2y ago
> However, with T steps of CoT, constant-depth transformers using constant-bit precision and O(logn) embedding size can solve any problem solvable by boolean circuits of size T There is a difference between being equivalent to a circuit
415.
by nyrikki 2y ago
ML is better than biological neurons in some tasks, they are different contexts. Almost all the performance of say college tests are purely from the pre-training, pattern finding and detection. Transformers are limited to DLOGTIME-uniform T
416.
by nyrikki 2y ago
Littlestone and Warmuth make the connection to compression in1986, which was later shown to be equivalent to VC dimensionally or PAC learnablilty. Look into DBScan, OPTICs for far closer lenses on how clustering works in modern ML commercia
417.
by nyrikki 2y ago
Here is a fairly good lecture series on graduate level complexity theory that will help understand parts. At least why multiple iterations help but why they also aren't the answer to super human results. https://youtube.com&
418.
by nyrikki 2y ago
Because you cite is about: > in-context learning LLMs have no concept of the symantic meaning of what they do, they just are dealing with next token prediction. "in-context learning" is the problem, not the solution to general
419.
by nyrikki 2y ago
The open domain frame problem is simply the halting problem. https://philarchive.org/rec/DIEEOT-2 While humans and computers both suffer from the frame problem, the LLMs do not have access to symantic properties, let a
420.
by nyrikki 2y ago
The major book sellers like Barns and Nobel carried them...even bought the OS/2 version of Borland C++ there.

More ›