Ot actually doesn't change their recommendation. There will be less, market share among people really worried about the American government. Others would be happy to pay for a better cloud run locally mostly under local laws.
Also, strong, random-looking passwords for droplets or apps saved in a text file. Use the Digital Ocean guide on setting up a Linux box securely and the UFW firewall. Then, lighttpd, BunnyCDN (esp for SSL), and periodic updates.
Works so well that it's easy to forget they're running.
I'll add many have remote access that is like a serial port. Digital Ocean has a "recovery console" that appears to work in a similar way. It runs in your browser. It works when other stuff breaks.
Now, for Linux, they have a SSH client in the browser enabled by default. I use it to manage a pile of $5/mo droplets, like the author.
I was bothered by heavy polarization of Americans, individually and even in churches, that appears to ve driven mostly by media outlets who cherry pick and lie. The Left and Right report specific events so differently that their readers might as well live in different worlds. People need to ditch those sources where possible. If not, they need to have a mix of them while understanding their biases.
Originally for churches, my draft article below describes how this problem affects all individuals and institutions. I recommend solutions which include AllSides.com (amazing!) and search engines for retrieving news from multiple outlets. I have a prototype. Progress is slow on my tool because I work two jobs with my free time mostly going to ministry serving Christ and others.
The ARINC scheduler, RTOS, and redundancy have been used in safety-critical for decades. ARINC to the 90's. Most safety-critical microkernels, like INTEGRITY-178B and LynxOS-178B, came with a layer for that.
Their redundancy architecture is interesting. I'd be curious of what innovations went into rad-hard fabrication, too. Sandia Secure Processor (aka Score) was a neat example of rad-hard, secure processors.
Their simulation systems might be helpful for others, too. We've seen more interest in that from FoundationDB to TigerBeetle.
We're already using domain-specific LLM's. The only LLM trained lawfully that I know of, KL3M, is also domain-specific. So, the title is already wrong.
Author is correct that intelligence is compounding. That's why domain-specific models are usually general models converted to domain-specific models by continued pretraining. Even general models, like H20's, have been improved by constraining them to domain-supporting, general knowledge in a second phase of pretraining. But, they're eventually domain specific.
Outside LLM's, I think most models are domain-specific: genetics, stock prices, ECG/EKG scans, transmission shifying, seismic, climate, etc. LLM's trying to do everything are an exception to the rule that most ML is domain-specific.
> We're already using domain-specific LLM's. The only LLM trained lawfully that I know of, KL3M, is also domain-specific. So, the title is already wrong.
This looks like an "ethical" LLM but not domain specific. What is the domain here?
> That's why domain-specific models are usually general models converted to domain-specific models by continued pretraining
I've also wondered this, like with the case of the Codex model. My hunch is that a good general model trumps a pretrained model by just adding an appropriate system prompt. Which is why even OpenAI sorta recommends using GPT-5.4 over any Codex model.
It's designed for drafting legal documents for lawyers. It's pretrained on a ton of court documents.
re why generalists are better
Much knowledge we have builds on prior knowledge. The prior knowledge is often reused across domains. Analogous reasoning, important in creativity, also connects facts or heuristics across different domains. Also, just being better at English.
If training a coding LLM, it needs to understand English, any concepts you type in, intrinsic knowledge about your problems, general heuristics for problem solving, and code with has comments and issues. The comments and issues might contain or need any of the above.
That's why I believe generalist LLM's further trained on code work better than LLM's trained only on code.
I was testing them on a HP laptop I bought for $200 with 4GB of RAM.
Windows, its default, used so much memory that there was not much left for apps.
Ubuntu used 500MB less than Windows in system monitor. I think it was still 1GB or more. It also appeared to run more slowly than it used to on older hardware.
Lubuntu used hundreds of MB less than Ubuntu. It could still run the same apps but had less features in UI (eg search). It ran lightening fast with more, simultaneous apps.
(Note: That laptop's Wifi card wouldn't work with any Linux using any technique I tried. Sadly, I had to ditch it.)
I also had Lubuntu on a 10+ year old Thinkpad with an i7 (2nd gen). It's been my daily machine for a long time. The newer, USB installers wouldn't work with it. While I can't recall the specifics, I finally found a way to load an Ubuntu-like interface or Ubuntu itself through the Lubuntu tech. It's now much slower but still lighter than default Ubuntu or Windows.
(Note: Lubuntu was much lighter and faster on a refurbished Dell laptop I tested it on, too.)
God blessed me recently by a person who outright gave me an Acer Nitro with a RTX and Windows. My next step is to figure out the safest way to dual boot Windows 11 and Linux for machine learning without destroying the existing filesystem or overshrinking it.
Consider a dedicated SSD for each OS. You should have a couple M2 slots in the laptop. What you can do is remove (or disable) the Windows SSD, install Linux on the second drive, and then add back the windows drive. Select the drive at startup you want to be in on boot and default the drive you want to spend most of your time in. I did that on my XPS and it was trouble free. Linux can mount your NTFS just fine, without having to consider it from a boot/grub perspective.
> Ubuntu used 500MB less than Windows in system monitor.
Those number meant nothing comparing across OS. Depends on how they counts shared memory and how aggressive it cache, they can feel very different.
The realistic benchmark would be open two large applications (e.g. chrome + firefox with youtube and facebook - to jack up the memory usage), switch between them, and see how it response switching different tasks.
Larger models better understand and reproduce what's in their training set.
For example, I used to get verbatim quotes and answers from copyrighted works when I used GPT-3.5. That's what clued me in to the copyright problem. Whereas, the smallest models often produced nonsense about the same topics. Because small models often produce nonsense.
You might need to do a new test each time to avoid your old ones being scraped into the training sets. Maybe a new one for each model produced after your last one. Totally unrelated to the last one, too.
reply