NewYorker "This is fine." pic.twitter.com/OiI3ynS3bT
Es erstaunt mich oft, dass Internetdienste ihre teuren Wordings nicht besser vor glanzraubenden Übersetzungen schützen.
Z.B. to Memorialize - in den Gedenkzustand versetzen (Facebook)
Kardinal knackt Stromkasten für besetztes Haus tagesspiegel.de/gesellschaft/p…
Word of the day: proto-italics
Source: reddit.com/r/linguistics/…
GDeyke Dopplergänger: looks like you as it approaches, still looks like you (but lower-pitched) as it passes you by.
nsaphra Something I've been working on as a reviewer lately: trying not to let the fluency and grammaticality of the language in a paper affect my judgment of the paper's clarity. A lot of L1 anglophones are really prejudiced on this. twitter.com/anas_ant/statu…
DrClaireH The moral of the story is, take care, oh university press offices, what you promote.
For what hath beene twote, yea but it cannot be untwote.
/fin
20/20
DrClaireH On the back of the Voynich Manuscript story, especially for those outside of academia, I wanted to shed some light on the issues that have allowed this story, and others like it, to happen. In the beginning, you submit an article to a journal relevant to your topic... 1/20
DrClaireH Sometimes, reviewers are utter bastards and they write heartbreakingly awful feedback that sends you face-first into a gallon-bucket of cheap icecream for the rest of the week. And for some reason, it seems to usually be Reviewer 2. I had one such just like this. 7/20
nlpmattg They had impact because they were easy to use. The ideas were shown to be general across a wide range of NLP problems, and they released tools that made actually using those ideas much easier.
nlpmattg What leads to adoption, then? I think it's almost entirely based on ease of use. If your idea is easy to adapt to other situations than you originally used it for, it will get adopted. If your tool is easy to use, people will use it.
languagejones Maybe the true meaning of the Voynich manuscript was friendship all along
Rule of thumb: the less Klingons and Vulcans are involved, the more interesting a Star Trek episode
ojahnn Hab heute während der Mittagspause mit einem begeisterten (+2 nicht so frohen) Kollegen den Themenkomplex "Vorfeldzersetzung" durchexerziert twitter.com/ojahnn/status/…
grumpwitch 16. Libraries remain the only place where you can spend hours in a publicly-accessible building without being expected to spend money. Parents come to entertain their children for free on wet days. People in poverty come for a warm place to sit. Libraries are a haven.
@nsaphra @wellformedness @kweenys Interesting! How are reviews evaluated? Simply on a point scale, or something more elaborate?
StefanMuelller @attac_hl @parabenstv @sciforfuture @FridayForFuture Der #Forschungsstreilk ist explizit dazu gedacht, die Erwachsenen mal wachzurütteln und eben vor den Universitäten präsent zu sein, so dass am 24.05. auch mehr StudentInnen und WissenschaftlerInnen kommen.
attac_hl @parabenstv @sciforfuture @FridayForFuture Der 24. Mai ist als weltweiter Generalstreik konzipiert. In Deutschland plakatiert Fridays For Future aktuell für diesen Tag überall mit der expliziten Aufforderung auch an Menschen jedes Alters.
gchrupala @nlpmattg I know it's been suggested before. But now it exists in the form of openreview.net/group?id=OpenR…, and some people (including me) already use it. We could now archive ALL *ACL submission on it, which would alleviate most of your problems with gatekeeping while preserving double-blind.
wellformedness Seeing a lot of takes about the scarcity of good reviewers, but absolutely nothing about compensating reviewers for their expert labor.
gchrupala (model => modest)
gchrupala I think we have a much better chance of success if we focus on coming up with smaller, more incremental changes and trialing them. My own model incremental proposal: all *ACL submissions are archived on an anonymous preprint server such as openreview.net.
gchrupala It is very unlikely that they will all agree to scrap it and adopt a new system overnight.
It is also unlikely that a very big and discontinuous jump in the design space will land at a better solution than the current one.
gchrupala The issue I have with these proposals is that they are too ambitious and too disruptive. There are tens of thousands of people and hundreds of institutions in dozens of countries invested in the current conference+journal system.
gchrupala Every couple of months someone comes up with a new Grand Proposal about how to fix the "broken" review and publishing model in NLP/ML. The latest one is from @nlpmattg.
labenzticker Eilau, die: Halsbrecherisches und verbotswidriges Überholmanöver, mit dem es besonders ausgebufften Autofahrern gelingt, 10 Sekunden früher an der nächsten roten Ampel einzutreffen. labenz.neutsch.org/Eilau
"Das ist rekursiv, da muss man ausgeschlafen sein."
DeletedWiki Third desk from the left in the second row from the back of Room 302, Bogstandard Junior School, Hamlet, Somerset
DeletedWiki Category: Mythological things at least in part based on chickens