demerphq

@chromatic_x Happy to help. It was a fun exercise, normally I don't hack that much on the optree. It was a good excuse to learn.

#perl 5.38.0 released. nntp.perl.org/group/perl.per… (thanks to @demerphq for doing the heavy lifting on the feature I started)

@briandfoy_perl I hacked something like that into core's Porting/sync-with-cpan recently. Sadly it isn't that easy for many cpan dists. I wish it were. It would be nice if there was a standard for Changes files.

@briandfoy_perl Try setting HARNESS_OPTIONS=j8 so that it runs tests in parallel.

When data modernization projects go horribly wrong: en.wikipedia.org/wiki/Burning_o…

@isotopp @mipsytipsy Or finding legacy codepaths that access a table that nobody realized exists... Or just finding dumb stuff. :-)

In terms of total observability, the stack frame is an underappreciated dimension. If the pipe is big enough sampling a tiny amount of data at this granularity is possible and revealing. Eg, we track db calls at the stack-frame level to find naughty code, @mipsytipsy @isotopp

@SCJOntario_en when will Judge Joseph Di Luca rule on the Dafonte Miller case? It has been almost four *years*. Justice delayed is justice denied. Covid is not an acceptable excuse, DO YOUR JOB. #BlackLivesMattters

@matseinarsen @lukasvermeer @TGuillerme @george_berry Nod. "False Loser" is a significant negative change in the underlying metric which is ALSO a Type-I error. False Winner/Loser are NOT replacements for the Type-I/II terms, they replace "false positive" and "false negative" when they are NOT used to refer to Type-I/II errors.

@demerphq @lukasvermeer @TGuillerme @george_berry Ah, I missed that Lukas meant loser for type I errors too, but I still stand by my point :)

@lukasvermeer @TGuillerme @george_berry FWIW, I am one of the non-experts who finds "False Positive" and "False Negative" confusing, but "Type-I" and "Type-II" unambiguous, if arbitrary. I like your proposed terms "false winner" and "false loser" for the two flavors of Type-I error. Hope they catch on.

@TGuillerme @george_berry Strongly disagree. In my experience, non-experts are likely to interpret the word “positive” as indicating directionality of the effect. A type-I (i.e. false positive) suggesting a negative effect (e.g. a drop in revenue) is often incorrectly referred to as a “false negative”.

@mipsytipsy Answering "When do I need observability?", is a bit like answering "When do I need my eyes?", the general answer is "Well, you don't strictly speaking *need* them. After all Blind people can do very well, but they sure make life MUCH easier."

"When do I need observability?" <= has begun to sound like "When do I NEED a lawn mower? The kids and I still have all our teeth, and they work just fine."

NYTimes reports that Shayla McCormally, lawyer for the Iowa Democratic party said "Any judgment of math miscalculations would insert personal opinion into the process". Which means her billable hours reports are too. Her clients will be pleased! Her accounting dept not so much.

@quimrovira God I feel old. I had to look that up. No. Thanks. :-)

@demerphq Should I send an onesie? 😜

@quimrovira Looks like I'll have to do the same @quimrovira. Just got the word that there was a case in a building attached to our Singapore office, and there was another case reported in my wifes parents neighborhood. Sigh.

@mipsytipsy @quimrovira IMO most people do alerts wrong. They send alerts to people, instead of having people subscribe to alerts. Once you have a central view you can do all sorts of interesting things.

oh no, oh no no no no no 😱 "drowning in alerts and can't understand your systems? pay lots of money for an AIOpsBot you don't understand. to filter, route, and escalate them down to a manageable roar!" blog.newrelic.com/engineering/wh…

@DJh4to @quimrovira I've lived in NL and DE (plus others). In NL they usually don't do anything on first contact with the doc. In DE they give you two weeks of antibiotics. :-) I think the NL approach is better, IF you know to be really insistent when you know that paracetamol won't cut it.

@quimrovira ah. hey being reluctant with painkillers is still better than giving strong and addictive painkillers. i am dutch and there are enough crazy things happening in the world that it no longer surprises me if they would do that, i would go. "yeh the world is fucking nuts."

sayeth xenu: basically when something is related to math and is useful you can be certain that it doesn't exist in c89

@quimrovira The Chinese authorities thought that too. Not anymore.

@demerphq Talking to se family in healthcare, they seem pretty convinced the overreaction is quite big. Oh well. Patience!

@demerphq I am wfh for the next two weeks, company policy, since my boyfriend just came back today from China. Tbh, in IT we are lucky that we can make things work fairly okay :)

Still no cases of #coronavirus in the Netherlands.. The aggressive policy to treat everyone arriving to Schiphol with paracetamol must be working.

Q: What kind of person throws a disabled man out of their wheelchair because he pointed out they were illegally parking in a disabled spot? A: A total and utter coward. #coward washingtonpost.com/nation/2020/01…