My second first-author paper! It took us way longer to write than I had expected, because despite the 4-sigma mass detection of the inner mini-neptune, we wanted to also statistically validate it just to be safe. (1/n)
Check out this paper from Lizhou Sha (@mentisoasis), me, Chelsea Huang, Dave Armstrong, @rafaelbrahm, and the @TESSatMIT team with the discovery of a hot Saturn with a nearby transiting planet! https://t.co/sseMfGk5Vl
@peteyreplies Another one of my favorite is to “table” a motion means opposite things in the US Congress vs UK Parliament (and other Westminster-style parliaments)
@AnthropicAI Why is @claudeai's @googlecalendar connector now requesting the broadest possible permissions with no opt out (w/o disabling entirely)? I don't want to grant "permanently delete all the calendars you can access", even if the connector doesn't allow Claude to do it.
⚠️New paper!⚠️ We have discovered a planet, Gaia-4b, and a brown dwarf, Gaia-5b, using the @ESAGaia spacecraft. Gaia-4b is Gaia's first planet detected with astrometry alone and represents the tip-of-the-iceberg of the 100s-1000s of planets that Gaia is expected to detect.
@GirishDuvvuri @YoniAstro ☝️ The fact that it's possible for the AASTeX LaTeX class to have a bug causing inconsistent line number output is evidence that there can't be a reliable programmatic way of detecting line numbers in the *output*, let alone whether the line nums are in the "main text".
@YoniAstro Not quite, because merely loading the lineno package does not guarantee the final PDF output will have line numbers. The only "easy" way I can think of is for arXiv to provide a modified version of lineno that reliably sets a variable whenever line numbers are toggled.
@YoniAstro If I have to guess, because there is no programmatic way of checking for the presence of line numbers in a LaTeX source file, they can only check for it in the compiled PDF with human moderators before announcement.
@alexteachey To be fair, the password prompt you're seeing is from the OpenSSH client, not GitHub. This is actually a security feature, because SSH servers that disables password authentication will *always* ask the client for passwords, so attackers won't know the difference.
@exoplaneteer Anyway, copyright is not the most suitable enforcement mechanism for lack of attribution (accidental or intentional) in academia, because academic ethics demand citation for *ideas* whereas copyright law only protects against verbatim copy of text/code.
@exoplaneteer But isn't attribution (by retaining the copyright notice) virtually the *only* requirement of the MIT license? Without it, it would not actually be a license at all.
@amvanderburg @exofastupdates @jcbastro ChatGPT evidently knows how to correctly split the advocate roles of Python and IDL between Smeagol and Gollum 🤣
@astro_noel On the extremely practical end, underrated but overpowered techniques like Numpy advanced indexing. Something like the official doc but more focused on examples and exercises https://t.co/JjhSAHsR2s
@virushuo I eventually threw my hand up at maintaining a Postfix instance and bit the bullet on ~Google Apps~ ~G Suite~ Google Workspace. It's enough of a hassle maintaining valid SPF, DKIM, DMARC, TLS-STS, TLSRPT configs already...