nqzero

what VM/provisioning are you using ?

!remindme 500 years

everything can be a moving part if you use enough hydrogen and oxygen

do you run vscode in the container ? if so, can you share your config ?

i've been trying to do this with systemd-nspawn

there are dozens of us !

i'm on a mid-level laptop, at times with slow or expensive internet, running ubuntu. i want to be able to run nominally-isolated "copies" of my laptop at near-native speed

1. each one should have it's own network config, eg so i can use wireguard or a vpn

2. gui pass-through to the host, eg wayland, for trusted tools, eg firefox, zoom or citrix

3. needs to be lightweight. eg gnome-boxes is dead simple to setup and run and it works, but the resource usage was noticeably higher than native

4. optional - more security is better (ie, i might run semi-untrusted software in one of them, eg from a github repo or npm), but i'm not expecting miracles and accept that escape is possible

5. optional - sharing disk with the host via COW would be nice, so i'd only need to install the env-specific packages, not the full OS

i'm currently working on a podman solution, and i believe that it will work (but rebuilding seems to hammer the network - i'm hoping i can tweak the layers to reduce this). does microsandbox offer any advantages for this use case ?

any idea how the webapp is written ? eg, what libraries or framework is used

this is 100% proprietary and the source code for this isn't available: https://frinklang.org/faq.html#OpenSource

i'm open to non-FOSS licenses (and advocate for and use a class of them), but in this case 1. afaict, the license offers zero rights to the user 2. for me personally, i want more rights for a language than for an app, eg mongo

they're trying to demo low-latency so they more-or-less have to be aggressive with cutting you off. that said, i think they're using filler to buy themselves a second or 2 - try a yes-or-no question

  - me: solo dev or small team
  - you: microsoft, mongo, redis

one of us is being disingenuous, but i don't believe it's me (username on point ;)

i agree with both your examples being bad. for #2, they required one-sided CLAs while "open source". the alternative to signing the CLA was to fork, which is rarely well-received by the community, ie the same basic issue i raised

FLOSS is great in that it can facilitate collaboration and adoption, but at the expense of greatly limiting the business models. and even then context still matters - eg there's big difference between the kernel with 1000s of independent contributors, and mongo with one party holding CLA rights to the entire codebase (FLOSS in name only, i'd argue)

  - for me, what's ultimately important is that people are free and that people that do good work are rewarded
  - software licenses are just a tool to help us get there
  - non-FLOSS might enable much of that same good while scaling to more business models and software
  - they might ultimately be good for society, but getting the details right is hard
  - i'm no longer actively working on my own stuff, but my attempt was: https://github.com/db4j/pupl
  - note: if i was doing this today, the core limit would be much higher

try releasing something under an open-but-not-open-source license as a solo developer or small team. there's a lot of established developers (presumably earning high salaries) that will very vocally badmouth the license choice. i'd seen this happen over and over again eg here, and when i've asked other developers why they open sourced their products they've said the same, and it was one of my concerns when i approached launch

sadly, my market fit was so bad that nobody ever looked at the license ::karma::

note: i have no problem with someone choosing not to use a product with a license they don't like (i do the same). it's the dissing of others that would use it that potentially crosses the line. i'm not even saying it *is* theft, only that there's a valid argument to that effect

is there an existing SLM that resembles an LLM in architecture that includes the code for training it ?

i realize the cost and time to train may be prohibitive and that quality on general english might be very limited, but is the code itself available ?

if you're funny, it's generally easier to work the wood

can you trademark a verb ?

amd a leg

maybe the board was just dogfooding ...

assuming that you have a potentially compromised linux host (or VM)

what level of client-side risk are you taking connecting a freeRDP client to that host ? (i'm not worried about the case of the client infecting the host)

what is the most secure similar application ? i prefer a linux client, though would be willing to consider windows if there was a client that was significantly more secure

there are new 2022-23 cell phones available for $50-70, including the cameras, battery, display, case, gps, DAC, usb, radio, accel ...

why aren't we seeing feature-rich dev boards based on the same platforms ?

> Hope no one sees/seizes this opportunity

time to start dogfooding ?

Is this English: Yes

(riffing off the "Is this Arabic" that was posted a few days ago - it's a lot easier in english, ie if the speaker says ti's english, it's english)

the Unidentified part is ok. but one plausible explanation of many of these is that they're purely optical/EM, ie there's no Flying Object

so at least until we're able to gather more data, UAP is much better

"generate large amounts of data quickly"

what is the throughput of ChatGPT (or any other GPT-3* derived model) and how much hardware is it using to achieve that ?

tangent: for a small air-to-air heat pump, what's the best commercially available coefficient of performance (COP) for a zero-delta-T ?

all the common ratings that i've found online refer to COP at a non-zero delta-t over a ranges of conditions. to illustrate my question, if you needed to pump heat from 70°F indoor air to 70°F outdoor air with a 1-ton unit, what would be the best COP you could achieve in 2022 ?

> I did not cherry pick. I just looked at each "interesting region" (interesting to me)

ok google, define "cherry pick"

i explicitly said "if the user executes ... they're powned" and never said anything about "running". you're implying i'm taking far more risk than i am

i'm trying to understand (and minimize, if needed) the automated risks of having untrusted files *stored* locally, which would give me time to read them and develop a level of trust

fwiw, if i need to run something untrusted, i'm using #2 some, but mostly:

  4. a 2nd (untrusted) machine running locally, which is beefier than my laptop and also used for benchmarking.
     i've never seen any unusual behavior from it, but treat it as though it's compromised

this thing is already more human than i am

certainly if the user executes or opens them (eg for a .doc) they're powned. but automated systems can also have exploits. i'm trying to make a list of these services (and maybe disable them) to minimize my footprint (often testing out untrusted code from github etc in a small secretive community, ie easy to target)

for ubuntu 21.04+, i'm aware of: - gnome-tracker-miner - gnome-thumbnailer (may require browsing in nautilus) - mlocate

at least the first two appear to be sandboxed, though unclear of the efficacy. any other services that you're aware of that would be automated vectors ?

> I would like to try writing a post before signing up

"signing up" is completely painless

  - no cost, no email required
  - choose a subdomain name aka login
  - choose a password
  - done

they've alredy given you exactly what you're asking for

unpopular opinion: null pointers (in at least java and c) are the single greatest metaphor in software development, and are the CS analog to the invention of zero

i've seen similar performance deltas from run to run for both jmh and non-jmh benchmarks, though i don't remember any of them being a 100% delta. very annoying because it means you need to wrap the benchmark in a bash script and average over a dozen+ invocations of java to get semi-meaningful results