This clip shows how we handle trace triage in https://t.co/oFxvKQQTuc:
stay on the list, open a side panel, inspect spans, check operational context, then jump to logs when needed.
Feedback welcome.
Most observability rollouts do not fail because the team is lazy.
They fail because the project starts with dashboard theology.
Week 1 becomes:
- Which charts
- Which team owns which board
- Which vendor-specific agent someone now has to babysit in prod
๐งต ->
No engineer has ever been paged because the pie chart palette was wrong.
The rollout that actually works is boring in the right way:
1. Create an API token
2. Point the collector
3. Land one trace, one log, one metric
4. Pick one workflow your team already hates
and then,
I share this story with founders of enterprise software companies, it's one of my favorite sales experiences and I think highlights some of the important non-technical aspects of sales:
I had just concluded a sales meeting with Very Large Company. This company had just publicly announced a huge commitment to Microsoft Azure.
The year is 2015 or so, and Azure is still very much an up-and-comer, so as the meeting concludes I ask the executive that was present "I'm curious, why Azure?" (As someone selling cloud agnostic tooling, I don't care one way or the other, but I was curious)
He says he'll show me on the way to the elevators. I consider that an odd answer to my question but go along with it.
As we near the turn to the elevators, we reach a hallway "intersection", and he points the opposite direction and says: "you see this hallway?" It was a standard corporate office hallway with small private offices flanking both sides as far as the building went. He continues, "Every single person in this hallway is a Microsoft paid employee."
And that's why they chose Azure. Because when something goes wrong, they don't have to submit a ticket or even call a phone number. They can literally knock on someone's door and get official, high tier support.
That's how Microsoft closed the 9 or 10 figure deal (multi-year, can't remember exact) to get this large company to go 100% into their cloud.
There's a lot of points to this story, but I think fundamentally the important thing is understanding the level of non-technical factors that lead to large enterprise deals.
I'll share one more story, this one a bit pithy but same vein, also coincidentally happened the same year:
I had just concluded a sales meeting with a Very Large Bank in London. Similarly, they had just announced a huge multi-year commitment to Azure. And similarly, I asked why. The CIO of the entire bank was present at the meeting. He leans forward and with a grin on his face says:
"Microsoft sent suits. Amazon sent hoodies. Google sent nobody."
10 years later, I still chuckle thinking about this interaction, and I still feel like despite the obvious humor, there's just so much underlying truth to it.
@boristane Hey Boris, unrelated, but the query builder on baselime has been broken for a while? While building a query (adding filters, visualisations, etc), I suspect it causes an internal infinite loop and crashes the app.
Live the product and the vibe. Thank you for building a great tool
Earlier today, our X/Twitter account @cheqd_io was compromised to link to a phishing website. This was an *incredibly* sophisticated social engineering/phishing attempt, where many of the usual red flags were hidden expertly. Wanted to share this as a cautionary tale... ๐งต
@mattpocockuk Absolutely love it. This comes in handy especially with something like Zod, just name the schme & type the same thing. Less import clutter plus you donโt have to do stuff like xyzSchema -> xyz
๐ Tadaaa ๐
โ vite-plugin-kit-routes is live & stable
After a few days on it, I'm pretty happy about the API now.
Everything inferred, pages, named actions, ...
Let's dig into it
๐งต๐ท๐