@diazona@techhub.social avatar

diazona

@diazona@techhub.social

Software engineer, former particle physicist, occasional blogger. I support the principle of cake.

This profile is from a federated server and may be incomplete. Browse more on the original instance.

matthewskelton, to random
@matthewskelton@mastodon.social avatar

"Surprisingly, our results show that women's contributions tend to be accepted more often than men's. However, when a woman's gender is identifiable, they are rejected more often. Our results suggest that although women on GitHub may be more competent overall, bias against them exists nonetheless."

https://www.researchgate.net/publication/308716997_Gender_bias_in_open_source_Pull_request_acceptance_of_women_versus_men

🤬

diazona,
@diazona@techhub.social avatar

@dahukanna @matthewskelton The paper says "We hypothesized that pull requests made by women are less likely to be accepted than those made by men. Prior work on gender bias in hiring [...] suggests that this hypothesis may be true." The way I read it, it sounds like that's why they were surprised: the results came out opposite to their hypothesis. Nothing to do with actual competence.

Plus, even though they didn't say so, I'd guess that bias against women in software engineering is discussed widely enough that the authors would have been aware of it and expecting to find that bias reflected in the rates of PR acceptance. I know not everyone is "tuned in" to the conversation about gender bias (and not everyone believes it 😠), but it seems unlikely that four people with the qualifications to write an academic paper on the topic would be unaware of it or in denial.

RickiTarr, to random
@RickiTarr@beige.party avatar

Today let's spread the love!

@ someone in the comments who has made your Mastodon experience better, a little explanation as to why would be awesome too! You are welcome to do more than one.

diazona,
@diazona@techhub.social avatar

@RickiTarr

The first one that comes to mind is @Taweret because it's day

Also @googlyeyesonmagiccards and @lowqualityfacts for sheer entertainment value

@covidsewage for making useful information a lot more convenient to access

@mcnees for the constant flow of interesting science facts

Honestly a lot of others too... the reason I'm enjoying the Mastodon experience is mostly the collective contributions of so many different people

andrew, to random
@andrew@esq.social avatar

IRS Direct File is a major win for the administration that doesn’t get a lot of attention.

Giving taxpayers an option to remove themselves from the marketplace of scammy for-profit preparation services is a major victory for all parties.

Other than the scammy for-profit preparation services, I guess.

https://www.forbes.com/sites/andrewleahey/2024/05/30/irs-direct-file-expands-to-all-50-states-for-2025/?sh=35669f5b825f

diazona,
@diazona@techhub.social avatar

@andrew woohoo! Very exciting. I never thought I'd see the day when this would happen.

(I think I'm over the income limit for it myself, but it's more about breaking the hold that the tax preparation industry has over the whole process)

leahawasser, to python
@leahawasser@fosstodon.org avatar

Hi friends. our second community created @pyOpenSci tutorial is out - How to make your code installable. making python code installable is the most minimal example of a python package and doesn't require any special tools! check it out here:

https://www.pyopensci.org/python-package-guide/tutorials/1-installable-code.html

fell free to leave any feedback for us or to ask questions in our @pyOpenSci discourse

https://pyopensci.discourse.group/c/coding-help/10

diazona,
@diazona@techhub.social avatar

@leahawasser @pyOpenSci Nice! I gave it a quick read-through, it's nice to see some good, modern advice out there (as far as I can tell, anyway) when it seems like so many other sources are giving outdated or bad advice about packaging.

diazona, to random
@diazona@techhub.social avatar

I think you could have guessed what happens when you leave the guy yelling "you can't do this, this is madness" as the only one in the control room where the madness can be started and stopped

brettcannon, to random
@brettcannon@fosstodon.org avatar

Announcing a proof-of-concept record type for Python https://snarky.ca/my-proof-of-concept-record-type/

If you happen to remember my struct syntax proposal, this is effectively implementing that idea via a function decorator. If there's enough positive feedback I will consider writing a PEP to propose this as syntax.

diazona,
@diazona@techhub.social avatar

@brettcannon Personally I wouldn't see myself using this new type. I mean, we already have named tuples and dataclasses which serve a similar role, and I already find the existence of those two types to be a source of uncomfortable ambiguity - not that they're the same, of course, but I often encounter situations where I could use either one and I have no good reason to choose one over the other, and the existence of record types as a third option usable in many of the same cases would only increase that ambiguity.

I suppose if other people find it useful, who am I to object - it's not really hurting me to have another type syntax that I never use - but I guess I just don't get it 🤷

#Python

diazona, to random
@diazona@techhub.social avatar

Tonight I am definitely looking up doing a seafood eating challenge

diazona, to random
@diazona@techhub.social avatar

So is it just a coincidence that we're watching Beyond the Time Barrier on the same day we passed a (Daylight Savings) Time barrier 🤯

diazona, to ObsidianMD
@diazona@techhub.social avatar

I started tinkering with @obsidian, and I'm definitely impressed - the experience seems smoother than any other notetaking app I've used. But I'm still debating whether it's enough better than @joplinapp (my current default) to overlook the fact that Obsidian is closed-source while Joplin is open.

It's not like I'm a die-hard open-source-only disciple, but all other things being equal-ish, I do prefer to choose open source programs, and I'm now asking myself just how far am I willing to stretch that "ish".

diazona, to random
@diazona@techhub.social avatar

"scientifically impossible" kind of went out the window when you discovered a tree stump with a heartbeat

diazona, to random
@diazona@techhub.social avatar

Super excited for a good old fashioned

I dunno what I was thinking last week, missing this to hang out with real people 🤦

otheorange_tag, to random
@otheorange_tag@mstdn.social avatar

Late to the party did they f-off into the sea .. of tranquillity?! --Dad

diazona,
@diazona@techhub.social avatar

@otheorange_tag @msh I'm glad somebody made a "fucking off into the Sea of Tranquility" joke, it needed to be said

I tried but wasn't quick enough and then there was a space wedding

diazona, to random
@diazona@techhub.social avatar

Ugh, it's one of those cases when I find myself contemplating which of Amazon or Walmart is less distasteful to do business with

Or eBay, but the thing I'm looking at sells for $60 on Amazon or Walmart and the various eBay sellers offering it are charging $130 minimum for comparable quality. I mean, I'm all for supporting smaller and (hopefully) less selfish sellers, but only within reason, and paying double the price is not kind of beyond reason in this case.

I wonder what others would do?

jacob, to random
@jacob@jacobian.org avatar

Every time I write about how I fucked something up with git, I get a good deal "this is your fault”-inflected feedback.

So look: I fuck something up with git about monthly. Have for a decade. You can draw one of two conclusions from this:

  1. I am terribly stupid.

  2. Git is fundamentally an unsafe system, to have been designed in such a way to allow for such repeated ongoing "mistakes”

If you think it's the former… why are you following such an idiot? The unfollow button is right over there!

diazona,
@diazona@techhub.social avatar

@meejah @jacob Yeah for sure. This is actually something I personally really like about Git, that when I mess up I basically always have a route to undo it, and I often don't find that to be true with systems that are simpler and more user-friendly.

diazona, to random
@diazona@techhub.social avatar

I'm rather lost by the movie and getting increasingly curious what the chickens are up to

diazona, to academia
@diazona@techhub.social avatar

Later this week I'm giving a colloquium to my old grad school department (physics) about my experience getting out of and working as a software engineer. It'd be interesting to crowd-source this: grad students and other former grad students of Mastodon, what would you want to hear in this kind of a talk?

hynek, to random
@hynek@mastodon.social avatar

Someone who didn’t suffer an outage due to mock-caused falsely passing tests hasn’t used mock long enough. https://fosstodon.org/@jankatins/111827209894488125

diazona,
@diazona@techhub.social avatar

@meejah @hynek @geofft Oh, I thought we were talking about situations where it's not possible to run the real service.

diazona, to random
@diazona@techhub.social avatar

What a classic today. I had gotten too used to extremely dumb movies so this was refreshing. Thanks as always @Taweret for hosting!

Next week should be excellent 🙂

meejah, (edited ) to random
@meejah@mastodon.social avatar

I'm producing JSON messages for consumption by other programs running this one as a subprocess. How should timestamps be communicated?

Aside: do non-browser JS runtimes also suffer from dumb limits (2**53)?

diazona,
@diazona@techhub.social avatar

@meejah @mkb It may be more portable than seconds since the epoch too...? I mean, I don't really know Windows programming at all but I don't think it uses the UNIX epoch, at least not natively.

And in fact seconds since the epoch isn't even a truly unique identifier of a time, because of leap seconds.

keithzg, to python
@keithzg@fediverse.keithzg.ca avatar

Was just thinking to myself that while language-specific package managers are a crime against God, at least ‘s pip is half-decent compared to others, but then I ran into a reason to use it and immediately:

RuntimeError: PyPI no longer supports ‘pip search’ (or XML-RPC search). Please use https://pypi.org/search (via a browser) instead.

and then eventually I got to the point where instead I was getting compile errors on the package in question somehow, and now I’m downgrading pip to “as much of an affront to nature as all the other language-specific package managers”. (Probably not even for the first time, I just forget in the meanwhile.)

diazona,
@diazona@techhub.social avatar

@meejah @keithzg Right, but I'd assume it's different subgroups within the PyPA that work on the two projects. I mean, surely it wasn't the pip developers, acting in that capacity, who made the decision to turn off search in PyPI.

diazona,
@diazona@techhub.social avatar

@keithzg @meejah I see... Well, I didn't exactly get the point, but your toot spawned an interesting discussion anyway.

I disagree with the premise though. I don't think language-specific package management is irreparably broken, nor do I think Python does it badly. I'm not saying it couldn't be better, but I have seen it be significantly worse.

diazona,
@diazona@techhub.social avatar

@meejah @keithzg I don't think it's splitting hairs. After all, if you complain to the pip developers that search doesn't work on PyPI (as I'm sure people have), what can they do about it, other than redirecting the question to the PyPI sysadmins? These distinctions are important to get complaints into the hands of people empowered to act on them.

It definitely is true that Python should have working search in its main package index, like other languages' ecosystems - but it seems like PyPI is operating under some constraints that I suspect other package indices are not. I hope they find a way to deal with it and can restore search functionality, but they do need time, and it hasn't been all that long since they turned it off.

moira, to random
@moira@mastodon.murkworks.net avatar

so ... xenophobia was the real monster all along?

diazona,
@diazona@techhub.social avatar

@moira Always is TBH

moira, to random
@moira@mastodon.murkworks.net avatar

sooooo... ummm... well... that was an... ending I guess...

seriously what the hell i like tokyo what even is happening here

there must be so much context i'm missing here

diazona,
@diazona@techhub.social avatar

@moira I wonder if it makes more sense after you eat the mushrooms

diazona, to random
@diazona@techhub.social avatar

The ants are very interested in these trees

I wonder what they're up to

diazona,
@diazona@techhub.social avatar

@moira oooh oooh wait there's an idea: a movie where the characters watch this movie and the ants climb out of the screen like in The Ring

  • All
  • Subscribed
  • Moderated
  • Favorites
  • •
  • provamag3
  • thenastyranch
  • magazineikmin
  • ethstaker
  • InstantRegret
  • tacticalgear
  • rosin
  • love
  • Youngstown
  • slotface
  • ngwrru68w68
  • kavyap
  • cubers
  • DreamBathrooms
  • megavids
  • mdbf
  • modclub
  • GTA5RPClips
  • normalnudes
  • khanakhh
  • everett
  • cisconetworking
  • osvaldo12
  • anitta
  • Leos
  • Durango
  • tester
  • JUstTest
  • All magazines
    •