4 ms·

It's funny to see this just now. I've started contributing quite a bit to Wikidata as part of some work that uses it as a data source, and it's really the first

It's funny to see this just now. I've started contributing quite a bit to Wikidata as part of some work that uses it as a data source, and it's really the first time in more than a decade that I work in an environment where my work gets public scrutiny. Shame is a really strong emotion for me, and I feel terrible when anyone spots a mistake. At one point, I left the site and didn't return for three days because I saw there was a notification for me. Which turned out to be positive feedback. I feel far worse about my own mistakes than about those of others. What I don't get is people doubling down on obvious mistakes. Show some contrition and your standing will net benefit from you screwing something up. Trust me on this: I'm German and we have made that principle semi-official government philosophy. Whenever I see, say, Turkish nationalists deny the Armenian genocide, or Polish wikipedia deleting articles about local anti-semitic incidents, I wonder if they seriously believe their actions won't make them look both guilty and somewhat stupid. Oh, while we're at it: Only by participating have I learnt what a vast enterprise the whole of Wikimedia actually is, and how almost all of it is open to the public. It's the only non-profit organisation at FAANG-scale (except Amazon I guess), and you might want to check out, for example, what Graphana looks like at scale: https://grafana.wikimedia.org/d/000000605/datacenter-global-overview?orgId=1 https://grafana.wikimedia.org/d/000000605/datacenter-global-...

shirakawasuna 6y ago
It's all about context. If you make an honest mistake on, say, WikiData, nobody should jump down your throat. That's not productive and the context is one where some degree of empathy should be expected. It's a commons. And trust me, I know how internet nerds can be on a commons - internet nerds that often get things very wrong in their criticisms but will swear and yell at you while leveling them. That context is important. The goal that I hope we all want to push for a data commons is a productive shared environment. But when it becomes narrative or editorials, now there are very serious political and scholarly issues at work and popular narratives are frequently self-sustaining via institutional norms. You can cite hundreds of documents and publications for a popular liberal (like lib dem) narrative even if it's no better supported than a position published by an esoteric scholar in Maine and win any arguments about which should be included in the definitive encyclopedia entry on the topic and which should be quietly shoved aside. Even disagreements listed in these pages tend to be relatively popular narratives that are not particularly disagreeable to those in power, or have become popular enough despite that opposition and are the exception. Turkish nationalists denying the Armenian genocide on Wikipedia will be drowned out by process: the dominant media and scholarly narratives (which are roughly correct). But what would we see on a page about Evo Morales back in October 2019? Would we not see the repeated dishonest narratives and headlines of the New York Times? In fact, we would: https://en.wikipedia.org/w/index.php?title=Evo_Morales&oldid=923888426 https://en.wikipedia.org/w/index.php?title=Evo_Morales&oldid.... No contextualization, just the implication that the NYT narratives are a good resource. And here is the closest snapshot in time to that wiki page: https://web.archive.org/web/20191112031842/https://www.nytimes.com/topic/person/evo-morales https://web.archive.org/web/20191112031842/https://www.nytim... In short, mistakes should always be contextualized by the roles of those involved, the goals sought (and hopefully shared), and a desire to get at the truth even if it's uncomfortable.

shirakawasuna 6y ago

It's all about context. If you make an honest mistake on, say, WikiData, nobody should jump down your throat. That's not productive and the context is one where some degree of empathy should be expected. It's a commons. And trust me, I know how internet nerds can be on a commons - internet nerds that often get things very wrong in their criticisms but will swear and yell at you while leveling them. That context is important. The goal that I hope we all want to push for a data commons is a productive shared environment. But when it becomes narrative or editorials, now there are very serious political and scholarly issues at work and popular narratives are frequently self-sustaining via institutional norms. You can cite hundreds of documents and publications for a popular liberal (like lib dem) narrative even if it's no better supported than a position published by an esoteric scholar in Maine and win any arguments about which should be included in the definitive encyclopedia entry on the topic and which should be quietly shoved aside. Even disagreements listed in these pages tend to be relatively popular narratives that are not particularly disagreeable to those in power, or have become popular enough despite that opposition and are the exception. Turkish nationalists denying the Armenian genocide on Wikipedia will be drowned out by process: the dominant media and scholarly narratives (which are roughly correct). But what would we see on a page about Evo Morales back in October 2019? Would we not see the repeated dishonest narratives and headlines of the New York Times? In fact, we would: https://en.wikipedia.org/w/index.php?title=Evo_Morales&oldid=923888426 https://en.wikipedia.org/w/index.php?title=Evo_Morales&oldid.... No contextualization, just the implication that the NYT narratives are a good resource. And here is the closest snapshot in time to that wiki page: https://web.archive.org/web/20191112031842/https://www.nytimes.com/topic/person/evo-morales https://web.archive.org/web/20191112031842/https://www.nytim... In short, mistakes should always be contextualized by the roles of those involved, the goals sought (and hopefully shared), and a desire to get at the truth even if it's uncomfortable.