Finally… After a decade of playing the Polluted Delta’s I got from Khan’s of Tarkir, I’ve upgraded to original Onslaught 👌
Next off… counter spell.
Daniel O’
- 11 Followers
- 62 Following
- 244 Posts
I’m just a machine that turns coffee in to bad code, bad Magic plays, and bad jokes.
For those out of the loop, I try to play with 100% original printing non-foil cards.
Partially cause clear communications (looking at you Inastrad basics) (https://scryfall.com/card/mid/270/island)
Partially cause I don’t feel the need to chase ever more shiny reprint cards
Gary "grim" KramlichRE: https://social.vmbrasseur.com/@vmbrasseur/115922995190919975
Everyone should speak their mind here. I wish I would haven't known about the donation request right after the survey as that feels tacky as hell because who the hell is going to donate to a project that's been disregarding its user's wishes for far too long.
Cat 🐈🥗 (D.Burch) 
cibyr
David Revoy
@8none1
On the topic of AI log analysis. I recently demo’s one of Elastic’s tools for this for work. In our context it was for security alerts.
Basically, the underlying data in the elastic index was already a vector database. So instead of getting the words out of the database, then running those through a language model. It was just looking for deviations and correlation in the vectors themselves. It was able to take a week’s worth of SOC logs and bundle them cleaning into a dozen real alerts. Then everything else could be marked as normal.
For a Linux sys admin, you’d probably be able to do something similar with a known-good vm image to do the vector statistics against.
On the topic of AI log analysis. I recently demo’s one of Elastic’s tools for this for work. In our context it was for security alerts.
Basically, the underlying data in the elastic index was already a vector database. So instead of getting the words out of the database, then running those through a language model. It was just looking for deviations and correlation in the vectors themselves. It was able to take a week’s worth of SOC logs and bundle them cleaning into a dozen real alerts. Then everything else could be marked as normal.
For a Linux sys admin, you’d probably be able to do something similar with a known-good vm image to do the vector statistics against.
Damn it!
Who let me run my VMs on a pool with no dram in the SSDs?!?
(VMs have been migrated now)
Who let me run my VMs on a pool with no dram in the SSDs?!?
(VMs have been migrated now)
ploum