we can time slice or do priviledge based, then it becomes a sharing issue.
we can time slice or do priviledge based, then it becomes a sharing issue.
ZLike 00:50 till you have a chance to run your snippet
Tcorrect. that usually isn't conducive to 60 minutes, though i know how much umesh loves the camera time
Ttoo complex; bad ux
Zwell if they are observing
Thmm
Tso, barring the wild security issues here, if the connection remains open for like, two days, you could have people reserve time. basically classic terminal.
Tuntil someone bricks it
ZGitHub
Is the five guest limit in place for technical reasons, or is it something that can be increased? I tried using it for a class that I'm running, and was disappointed that most people couldn'...
Zso seems vs live code is set to 30
Zexactly
ZBut this is active participants, not observing participants
Zin theory that upper limit is unreachable
Twhat's the hw target here i can't tell
Tit's just vs code
Ti thought we were talking about deploying code... at least in your 32-bit savior example
Zyes if he can type on my ide and my stm32 is conneceted
Zhe can deploy
Zhe has access to the toolbar and the cli sections and thats what he needs to deploy
TOkay, then. What about:
- Security issue
- Device bricking
TI bet @Ming has some useful experience to contribute here
ZSecurity I suspect its a walled garden thing, the subject machine will just be used to for this purpose and will persist nothing
TYes, will have to be sandboxed; same thing with the safety of the platform. I just worry we end up partitioning out the utility of it
ZDevice Bricking might be more difficult maybe we do a userland / kernel type separarion
Zlikely not
T@Umesh Lokhande do you see this being useful
Zbecause the ideal person wants to try this and if it sort of works they want to go back to their IDE and keep on going
Texcept for the fact they probably don't have the hardware... which is the point
ZThat's true, lots of moving parts. However all this is possible.
TMaybe distributor friends would be willing to lend a discount. Everyone who wanted to do it though would probably have to be a pro and provide info
Tbut, it could work
TWho would manage this infrastructure on DH side? @ZacckOsiemo? @Phoenix?
TWhat do we need to support this at high rel
Pwe could manage it, For high reliability, we’d need a sacalable cloud infrastructure to handle increased traffic during live coding sessions. AWS or DigitalOcean could be used for this.
PLoad balancing and redundancy to ensure minimal downtime and consistent performance.
TWhat would that do to our costs
Pcosts will increase based on usage. More users or live events = higher costs for compute power, storage, and bandwidth. Estimation depends on traffic, might see costs start at $100–$500/month for small-scale usage and increase to $1000+ for higher traffic or events.
Zwhat VPS are you using?
PAWS because it offers autoscaling, so you’ll pay more during high traffic, but it’s managed well.
Tyeah sry i should have said ~50 concurrent users
ZI would start with a 5 $ linode and go until it explodes
ZWe dont need autoscaling, it's too ealy
Pwe don't want stuff to explode mid event
ZOh those are the best ones
Pvery bad ux
Zvery big news though
Pemotional damage
Tdifferent kind of explosion
Z
T
T^^Umesh during his next Engineering Hour
