eb0la

39 ms·

421.
by eb0la 7y ago
I guess It's about liability and reproducibility. When someone does harm with his/her car, holds the civil (and sometimes penal) responsability of his own acts and usually cannot do harm more than once in a very small timeframe.
422.
by eb0la 7y ago
I guess It could be useful to automate scripts. Instead of having expect waiting for a prompt, just wait for 'Bye' and start the next step.
423.
VistaVision(en.wikipedia.org)
1 points by eb0la 7y ago|0 comments
424.
by eb0la 7y ago
Of course they can co-exist :-) ! I guess I should write am essay about _when_ one is better than the other... just to see FaaS catch up in short, short, time :-D
425.
by eb0la 7y ago
Mostly containers with K8s and Istio. Works good but it's complex to start (for my taste). For data we still have VMs with Spark or traditional ELTs like Talend (depending on the customer); but are moving forward to Apache Beam running
426.
by eb0la 7y ago
I really like the acronyml ( Plonk = Prometheus + Linux/Linkerd + OpenFaas + Nats + Kubenetes ) I miss some long-running ETL/dataflow pipeline-like service like Apache Beam or something similar. The reason is I don't like to
427.
by eb0la 7y ago
Not Dutch; but gas and electrical companies are also a reckless mafia in my country (Spain), so I had to upvote. Dude, I wish you the best.
428.
by eb0la 7y ago
Maxmind GeoIp database. If you plan to use it, remember to pay forma it and upgrade every 2-5 days. If I remember well, about 1% of the results were not up to date after 7 days. And the free database was at least 2-3 weeks old.
429.
by eb0la 7y ago
The market for GCP architects and data engineers is quite hot for now. K8s might be a game changer for data in the next months: - Spark can be run inside k8s. - There are some "spark replacements" in development that get routinely
430.
by eb0la 7y ago
My case was not Big Tech, but Big Telco and the problem was a bit similar, since doing anything involved a pay cut. First, I took _a long_ time off to rethink my career. I was really lucky to have the time, money, and support from my wife
431.
by eb0la 7y ago
If you purchase a Carepaq, can actually extend the service beyond that? 3,4,and 5-Year 24x7 Carepaqs are available on purchase for the _whole_ chassis and parts... but I wonder if I can keep purchasing 1-year carepaq extensions beyond the 6
432.
by eb0la 7y ago
ARIN (North America) and APNIC (Asia-Pacific) ran out of addresses some time ago (2015 I guess). Game over IPv4.
433.
by eb0la 7y ago
Inceptionv3 and Resnet50 - for detecting cracks in displays and SKUs in a supermarket. And a Flask webservice for serving. I wished I could use Jupyter Serve; but Flask was easier for our ops team.
434.
by eb0la 7y ago
Another patterns: - $(SHINY_NEW_TECH) [beats|as a better|replaces] $(OLD_TECH) - $(OLD_OR_BORING_TECH) beats $(SHINY_NEW_TECH) (because) - $(OLD__TECH) refuses to die
435.
by eb0la 7y ago
"$(PropietarySoftwareCO) [is now using|adopts|champions] $(OpenSourceTechnology)" This works because there is an implict conflict.
436.
by eb0la 7y ago
I'm also a heavy fountain pen user and I've found Parker ink (the same my father used), dries fast for fine-sized nibs. The black-blue looks really good on paper. Currently using a Waterman fountain pen, but also have a Lamy safar
437.
by eb0la 7y ago
Perl is #1 in Advent (perladvent.org). I've been following Perl Advent for a looong time. The sad art is I don't code Perl anymore (at least 10-12 years) but I still visit perl advent every year. Let's see what happens this y
438.
by eb0la 7y ago
Google has Anthos (kubernetes on prem, with access to GCP marketplace).
439.
by eb0la 7y ago
I always enjoyed reading Charles Petzold code on Microsoft Systems Journal. When I discovered the Magazine (back un the 90's) I wasn't ready to code for windows nor win32 because I was a broke student with a 1MB 80286; but I am pr
440.
by eb0la 7y ago
I believe the smart part is he knows what he wants... and that's hard, btw.
441.
by eb0la 7y ago
Remember web counters for visits? My company started them always at 1000 :-D
442.
by eb0la 7y ago
Some accounting firms actually take this into account when you're sending bills. In my country bills must have an auto incremented bill number. You cannot issue bill #100 without having 99 bills before. I know one auditing firm that us
443.
by eb0la 7y ago
I always thought it is more like making a building without a scaffold: when you're on the ground floor it doesn't look bad, but the more you build....
444.
by eb0la 7y ago
If you are doing a serious ETL job, you'll need to handle rejects - I mean, a way to identify which rows of data cannot be handled gracefukky by your ETL. Doing this in plain SQL or Spark is complicated. This is usually a job for comme
445.
by eb0la 7y ago
Also great for BI. Most BI tools make a lot of calculations in their queries just to get the data as they want. With this, you can get that calculations pre-made and persisted except when you make a backup.
446.
by eb0la 7y ago
Try the first $20 million is the hardest, ir the nudist un the night shift by PO Bronson. They are both about the dotxom boom: the first one is a novel about java. The second is journalism (if I remember well).
447.
by eb0la 7y ago
With cargo (Rust build tool) you don't need to use crates.io _always_ . For instance, you can target a repository and even a specific commit like this in gour cargo.toml file: [dependencies] package = { git = "https:/&
448.
by eb0la 7y ago
Reminds me Tiddlywiki ( https://tiddlywiki.com/); but with spreadsheets instead of wikis... ... as a second thought spreadsheets are the wikis of people in finance, management...
449.
by eb0la 7y ago
Good Idea? yes and no. The biggest problem I find is makes expectations _high_ from the beginning, which is a bad idea. Most data projects are about managing expectations. Also, it is _very_ hard to demonstrate that your model works if the
450.
by eb0la 7y ago
Tobacco, etc. Because life is too short to think you're helping people become sick.

More ›