Kshitij Aranke @rn.ke · 05/02/2025is it just me, or does this comparison make anyone else uneasy? 140
Kshitij Aranke @rn.ke · 16/01/2025youtu.be/kOO31qFmi9A?...youtu.beIntroduction to Microsoft Excel 1990YouTube video by Jan Brehm 031
Kshitij Aranke @rn.ke · 16/01/2025what does the map of your data look like e.g. Harry Beck optimized for density and legibility over geographical accuracy in the map below 030
Kshitij Aranke @rn.ke · 16/01/2025and go the other way for dimensional modeling: 1. facts are events/verbs 2. dims are objects/nouns 030
Kshitij Aranke @rn.ke · 14/01/2025I haven't heard back either. Last year, I got the rejection email on Jan 24. 100
Kshitij Aranke @rn.ke · 14/01/2025I find these few extra minutes of work a quick win to start/end the day, so would prefer to adjust tickets manually. 010
Kshitij Aranke @rn.ke · 14/01/2025Right, we do this via automations. * If you put the ticket number in the PR description, it auto-links to the issue. * Once PR is merged, issue is also closed. 110
Kshitij Aranke @rn.ke · 14/01/2025Never tried it, and never heard of anyone using it either. I’ve never needed integration between a ticket tracker and my IDE, can always pull up a browser window. 100
Kshitij Aranke @rn.ke · 14/01/2025It’s been around for 15 years, not a new thing. en.wikipedia.org/wiki/YouTrac... 100
Kshitij Aranke @rn.ke · 10/01/2025Highly recommend doing this for the speaker training sessions by themselves, worth their weight in gold. 020
Kshitij Aranke @rn.ke · 03/01/2025Is there a good write-up of how last-touch attribution won? Asking in context of recent Honey scandal #databs 120
Kshitij Aranke @rn.ke · 22/12/2024cool, wanna give us a hand with these? /s github.com/dbt-labs/dbt... 100
Kshitij Aranke @rn.ke · 19/12/2024I don't work at a transit agency, but doing a "transit wrapped" seems like it would be fun. The basic idea is you upload a csv with all your trips and a Shiny/Quarto dashboard shares interesting stats. 120
Reposted by Kshitij Arankedietrich @burrito.space · 19/12/2024deeply feeling RFC 6919 as of late datatracker.ietf.org/doc/html/rfc... 2223
Kshitij Aranke @rn.ke · 19/12/2024€120 with taxes and shipping seems...reasonable? At least more reasonable than buying the Supreme brick for $30 in 2016? 000
Kshitij Aranke @rn.ke · 18/12/2024we needed a white-collar alternative to the makita coffee maker youtu.be/4xOEIpbxM4w?...youtu.beThe Makita Coffee Machine: A Bizarre Battery-Powered BrewerYouTube video by James Hoffmann 021
Kshitij Aranke @rn.ke · 14/12/2024i'm more surprised that they haven't been on the internet long enough to know the usual tricks of how to get around a paywall 010
Kshitij Aranke @rn.ke · 14/12/2024I’m not asking for free, but I do want a middle ground where I can buy access to an article a-la-carte or as a prepaid bundle across publications. Needing a subscription to read a single article is madness. 110
Kshitij Aranke @rn.ke · 11/12/2024Yeah, both approaches are O(n^2) and get you to the same result 😄 The iterative approach I took is just more intuitive to me because I write Python not SQL all day 110
Kshitij Aranke @rn.ke · 11/12/2024Wait I found a solution in pure SQL using list transformations (works in DuckDB): gist.github.com/aranke/74206... 161
Kshitij Aranke @rn.ke · 11/12/2024I think there is one if you "cheat" by using dbt Jinja/UDFs: 1. Iterate over dates to create a list of cooldown dates (date - cooldown_dates[-1] > 90) 2. If date in cooldown dates, return true else false 120
Kshitij Aranke @rn.ke · 10/12/2024can you expand on this? as an uninformed outsider, is calling the go/rust sdks not feasible? 110
Kshitij Aranke @rn.ke · 04/12/2024Or maybe you still need it but it's abstracted away from you so you might not even realize you're using it? 130
Kshitij Aranke @rn.ke · 04/12/2024Is E the killer feature in ETL/ELT for Iceberg or is it more T/L? 100
Kshitij Aranke @rn.ke · 04/12/2024Is it just me, or does this feel really risky? Hypothesis looks cleaner to me 100
