#cofacts
2021-05-02
⚠️ *Action required - upgrade app for g0v* ⚠️
@lucien hello. nice to meet you. We are in Tokyo and we would like to build cofacts japanese version + machine learning fake news scoring model. I know this is open source but who is the leader of this? Hope we can help other local languages as well
- 👍1
Hi. It’s glad to hear you want to build Japanese cofacts! We actually have a Thailand fork one.
@mrorz is the founder of cofacts, too.
- 👍2
- 🍣1
This message was deleted.
I think this conversation already take place here in #cofacts 😆
I think @bil has a Japanese version of Cofacts website, haven’t have the chance to put it back to our repository
For the time being please refer to the English version here: https://en.cofacts.tw/
en.cofacts.tw
Cofacts is a collaborative system connecting instant messages and fact-check reports or different opinions together. It's a grass-root effort fighting mis/disinformation in Taiwan.
I think this conversation already take place here in #cofacts 😆
@mrorz @lucien Thank you! We will check
@john989 hey! pls see above. Let’s talk with Bil
@bil Hi! こんにちは！Have you done Japanese version of cofacts? We want to contribute to it
thank you! I read this. so have you deployed this??
@dai I think I can integrate this into our codebase. Let me try it out for now
wow. thank you
Related pull requests (description WIP)
• rumors-site - https://github.com/cofacts/rumors-site/pull/432
• rumors-line-bot - https://github.com/cofacts/rumors-line-bot/pull/251
• rumors-site - https://github.com/cofacts/rumors-site/pull/432
• rumors-line-bot - https://github.com/cofacts/rumors-line-bot/pull/251
@mrorz Hi! Your Thailand project is fantastic. Do you know any other country who needs cofacts? We can provide our brand new translation API for free for such project
Thanks! All kudos should go to OpenDream when it comes to http://cofact.org/ (Cofact in Thai). Cofacts working group here in Taiwan just had one or two meetings with them and they managed to sort out all the things.
I think starting a local fact-checking community is the most important and difficult part to do when it comes to crowd-sourced fact-checking. We are glad to provide our source code to anyone that want to start such a community in their own country, so that they can focus on community building more quickly.
I think starting a local fact-checking community is the most important and difficult part to do when it comes to crowd-sourced fact-checking. We are glad to provide our source code to anyone that want to start such a community in their own country, so that they can focus on community building more quickly.
Developer portal of Cofacts: https://beta.hackfoldr.org/cofacts/https%253A%252F%252Fhackmd.io%252Fs%252Fr1nfwTrgM
Run Cofacts on laptop: https://beta.hackfoldr.org/cofacts/https%253A%252F%252Fg0v.hackmd.io%252Fs%252FSJ1f6uU5X
Cofacts Toolkit - User manuals, tutorials, policies, etc: https://beta.hackfoldr.org/cofacts/https%253A%252F%252Fhackmd.io%252Fc%252FSklM4dV9m%252F
Cofacts analytics (usage): https://cofacts.g0v.tw/analytics
Cofacts editor’s site (english): https://en.cofacts.tw
Cofacts LINE bot (english): https://lin.ee/2Ox46aqyW
Run Cofacts on laptop: https://beta.hackfoldr.org/cofacts/https%253A%252F%252Fg0v.hackmd.io%252Fs%252FSJ1f6uU5X
Cofacts Toolkit - User manuals, tutorials, policies, etc: https://beta.hackfoldr.org/cofacts/https%253A%252F%252Fhackmd.io%252Fc%252FSklM4dV9m%252F
Cofacts analytics (usage): https://cofacts.g0v.tw/analytics
Cofacts editor’s site (english): https://en.cofacts.tw
Cofacts LINE bot (english): https://lin.ee/2Ox46aqyW
1
Replied to a thread: 2021-05-02 21:16:10
Thanks! All kudos should go to OpenDream when it comes to http://cofact.org/ (Cofact in Thai). Cofacts working group here in Taiwan just had one or two meetings with them and they managed to sort out all the things.
I think starting a local fact-checking community is the most important and difficult part to do when it comes to crowd-sourced fact-checking. We are glad to provide our source code to anyone that want to start such a community in their own country, so that they can focus on community building more quickly.
I think starting a local fact-checking community is the most important and difficult part to do when it comes to crowd-sourced fact-checking. We are glad to provide our source code to anyone that want to start such a community in their own country, so that they can focus on community building more quickly.
cofact.org
คนใกล้ชิดของคุณ อาจตกเป็นเหยื่อของข่าวลวง หรือ ส่งต่อข่าวลวงบนอินเทอร์เน็ตโดยไม่รู้ตัว
1- 1
thank you! I read this. so have you deployed this??
@dai I think I can integrate this into our codebase. Let me try it out for now
wow. thank you
@john989 let’s catch up
@mrorz this is our thought. of course we will try to adjust our actions towards your direction.
YouTube
@ggm I think you can add your effort regarding NLP here.
@john989
Currently in Cofacts we use NLP just for categorizing the message into different topics, hoping that this can be useful to the volunteer editors. Personally I think it would be very difficult, if possible, that we can directly use NLP to infer correctness and generate reply, since most messages actually require quite some human knowledge to answer correctly. To make matters worse, almost all suspicious messages requires editors to sense the emotion (fear/anger/hatred/etc) in the message that makes the message viral, and provide precise and short answers to “defuse” that emotion.
In our previous research ( https://hackmd.io/@ggm/HkhTS45ig?type=view ) there were some model that determines if a viral message is disinformation by looking at the pattern it spread in the social media; however this method is not possible in closed messaging apps like LINE and whatsapp.
@john989
Currently in Cofacts we use NLP just for categorizing the message into different topics, hoping that this can be useful to the volunteer editors. Personally I think it would be very difficult, if possible, that we can directly use NLP to infer correctness and generate reply, since most messages actually require quite some human knowledge to answer correctly. To make matters worse, almost all suspicious messages requires editors to sense the emotion (fear/anger/hatred/etc) in the message that makes the message viral, and provide precise and short answers to “defuse” that emotion.
In our previous research ( https://hackmd.io/@ggm/HkhTS45ig?type=view ) there were some model that determines if a viral message is disinformation by looking at the pattern it spread in the social media; however this method is not possible in closed messaging apps like LINE and whatsapp.
Thanks! I’ll check the article out really interested to take a look into this research.
Wondering if there are models that could be used to detect strongly emotional messages that would be typically associated with fake information
Wondering if there are models that could be used to detect strongly emotional messages that would be typically associated with fake information
As far as I know Cofacts never tried to detect emotion in rumor text.
Institute for Information Industry do develop such feature in Rumor Catcher ( https://tfc-taiwan.org.tw/articles/4477 ). However, they are not a fan of open-source and want to hold their findings proprietary.
Institute for Information Industry do develop such feature in Rumor Catcher ( https://tfc-taiwan.org.tw/articles/4477 ). However, they are not a fan of open-source and want to hold their findings proprietary.
Not sure if they published any academic papers regarding this issue. I think they should have 😛
Another direction may be
https://www.niemanlab.org/2020/07/a-lesson-in-automated-journalism-bring-back-the-humans/
https://www.niemanlab.org/2020/07/a-lesson-in-automated-journalism-bring-back-the-humans/
這也太難找，雖然說一聽就不是英文但是不知道怎麼查證這是在哪裡
Replied to a thread: 2021-05-02 21:39:34
@ggm I think you can add your effort regarding NLP here.
@john989
Currently in Cofacts we use NLP just for categorizing the message into different topics, hoping that this can be useful to the volunteer editors. Personally I think it would be very difficult, if possible, that we can directly use NLP to infer correctness and generate reply, since most messages actually require quite some human knowledge to answer correctly. To make matters worse, almost all suspicious messages requires editors to sense the emotion (fear/anger/hatred/etc) in the message that makes the message viral, and provide precise and short answers to “defuse” that emotion.
In our previous research ( https://hackmd.io/@ggm/HkhTS45ig?type=view ) there were some model that determines if a viral message is disinformation by looking at the pattern it spread in the social media; however this method is not possible in closed messaging apps like LINE and whatsapp.
@john989
Currently in Cofacts we use NLP just for categorizing the message into different topics, hoping that this can be useful to the volunteer editors. Personally I think it would be very difficult, if possible, that we can directly use NLP to infer correctness and generate reply, since most messages actually require quite some human knowledge to answer correctly. To make matters worse, almost all suspicious messages requires editors to sense the emotion (fear/anger/hatred/etc) in the message that makes the message viral, and provide precise and short answers to “defuse” that emotion.
In our previous research ( https://hackmd.io/@ggm/HkhTS45ig?type=view ) there were some model that determines if a viral message is disinformation by looking at the pattern it spread in the social media; however this method is not possible in closed messaging apps like LINE and whatsapp.
HackMD
【真的假的】其他相關新聞 ====== 2021-03-24 研究：2020總統大選前 候選人粉專藏2成假帳號 <https://www.cna.com.tw/news/firstnews/202>
mygopen 真的超猛
我其實有試過手機開 google translate app
自己瞎選語言、按下麥克風按鈕
並且把手機靠近電腦喇叭然後播放
但我選的每個語言他都沒有辨識出來 QQ
自己瞎選語言、按下麥克風按鈕
並且把手機靠近電腦喇叭然後播放
但我選的每個語言他都沒有辨識出來 QQ
Cofacts 在等一下（0:00）要重啟 nginx 來試著放上日文版網頁，希望只會斷線 30 秒，但如果設定有問題的話可能會壞掉久一點 QQ
Hi all, we are going to restart nginx at 0:00 to get japanese version of Cofacts online. Hopefully it will only take 30 seconds, but may be longer if things go south.
Hi all, we are going to restart nginx at 0:00 to get japanese version of Cofacts online. Hopefully it will only take 30 seconds, but may be longer if things go south.
- 🙏2
2021-05-03
Hmm cofacts.tw is still alive, which is good. ja.cofacts.tw needs SSL setup on Cloudflare.
Japanese version of Cofacts website UI is up and running now:
https://ja.cofacts.tw/
Note that the database is still connected to our production server, which serves Mandarin content, thus ja.cofacts.tw is just for demonstration purpose.
I am putting together a pull request that include the current Japanese translations so that after merging the pull request, @john989 and @dai can fork the project with Japanese translation already in place and build your own version using this as a starting point 🏗️
https://ja.cofacts.tw/
Note that the database is still connected to our production server, which serves Mandarin content, thus ja.cofacts.tw is just for demonstration purpose.
I am putting together a pull request that include the current Japanese translations so that after merging the pull request, @john989 and @dai can fork the project with Japanese translation already in place and build your own version using this as a starting point 🏗️
ja.cofacts.tw
「Cofacts デマ探偵隊」はインターネット情報及びデマ検証・報告に関する市民協働とリンクするシステムです。デマ情報に対して、一人一人が積極的に参加するような草の根の力で対応策を作り上げるのを目指しています。
- 🙏1
• Include Japanese translation in codebase • Build & push script • Relax landing page style so that it also works for Japanese version
*Pull Request Test Coverage Report for <https://coveralls.io/builds/39290812|Build 1315>* • *0* of *0* changed or added relevant lines in *0* files are covered. • No unchanged relevant lines lost coverage. • Overall coverage remained the same at *86.629%* * * * * * * *:yellow_heart: - <https://coveralls.io|Coveralls>*
JP version of Cofacts chatbot:
https://lin.ee/hIyZFR8
https://lin.ee/hIyZFR8
m.facebook.com
【聊聊產品遊戲化中的雙循環理論】 「遊戲化」到底是不是一個偽命題一直有些爭論。贊成者認為遊戲中對於人性誘因的設計，適用於各種產品如拼多多、Dulingo、Orangetheory Fitness等都有不錯的成績；反對者則認為很多打著遊戲化的產品大多流於表面，短期內可能帶來不錯的成效，但長線看來對產品的收入沒有太大幫助。...
- 😂1
Related pull requests (description WIP)
• rumors-site - https://github.com/cofacts/rumors-site/pull/432
• rumors-line-bot - https://github.com/cofacts/rumors-line-bot/pull/251
• rumors-site - https://github.com/cofacts/rumors-site/pull/432
• rumors-line-bot - https://github.com/cofacts/rumors-line-bot/pull/251
Thanks! I’ll check the article out really interested to take a look into this research.
Wondering if there are models that could be used to detect strongly emotional messages that would be typically associated with fake information
Wondering if there are models that could be used to detect strongly emotional messages that would be typically associated with fake information
@john989 this is Japanese police info about fake info
@john989 you can find some typical features
@mrorz please google translate this
Thanks for the info!
Maybe we cannot assert the message’s correctness by these emotion traits, but I think it would be helpful to mark out the sentiments in the rumor text and ask the audience to be careful, or just raise awareness that such message is trying to sway you by emotion.
Maybe we cannot assert the message’s correctness by these emotion traits, but I think it would be helpful to mark out the sentiments in the rumor text and ask the audience to be careful, or just raise awareness that such message is trying to sway you by emotion.
As far as I know Cofacts never tried to detect emotion in rumor text.
Institute for Information Industry do develop such feature in Rumor Catcher ( https://tfc-taiwan.org.tw/articles/4477 ). However, they are not a fan of open-source and want to hold their findings proprietary.
Institute for Information Industry do develop such feature in Rumor Catcher ( https://tfc-taiwan.org.tw/articles/4477 ). However, they are not a fan of open-source and want to hold their findings proprietary.
Not sure if they published any academic papers regarding this issue. I think they should have 😛
Another direction may be
https://www.niemanlab.org/2020/07/a-lesson-in-automated-journalism-bring-back-the-humans/
https://www.niemanlab.org/2020/07/a-lesson-in-automated-journalism-bring-back-the-humans/
```There are many expressions that emphasize, arouse anxiety, and hurry.
If the information is true, it is sufficient to include only the facts. However, false information is made to believe in emphasized expressions and expressions that arouse anxiety, and tries to spread it in a hurry.
Examples of expressions that are easily included in hoaxes
Emphasized expression Very, quite, very, absolutely, all, surely, serious, etc.
Expressions that arouse anxiety Dangerous, malicious, difficult, etc.
Expression to hurry Immediately, urgently, etc.
Content related to life and money
Many life-threatening contents such as evacuation information and disaster information are also abused. In addition, money-related items such as "free services will be charged from next month" and "paid services will be free for a limited time" are also easily abused, so be sure to check the information source.
Source not listed
If it's true, but it doesn't have a link (URL of the source) or rationale, be sure to search for it yourself to find out the source. If it is related to government offices or companies, please check the official website or official blog.
It is also effective to check TV, radio, etc. depending on the content such as weather information.
Written in hearsay format
You should also be careful if the information provided includes hearsay formats such as "likely," "like," and "likely." If the source of information is ambiguous, such as "friends", "acquaintances", or "entertainers", please check with the parties directly or by searching the official website.
Recommended to spread
There are two main types of writing that recommend diffusion: malicious and good faith. In either case, it's important to know the source and not spread indiscriminately.
Examples of malicious spread demands
"If you don't give it to someone within ○ hours after seeing it, you will be unhappy."
"People who stop this information will be identified and attacked (captured)" etc.
An example of a request for spread in good faith
"The pet shop has collapsed, so please spread it to people who are likely to get it."
"Please share it with all your acquaintances to prevent damage" etc.
In other special cases, the source of information itself may be a false site (so-called joke site). Some people believe in content that is not common sense. If you receive such information, get in the habit of verifying the authenticity of the information.```
If the information is true, it is sufficient to include only the facts. However, false information is made to believe in emphasized expressions and expressions that arouse anxiety, and tries to spread it in a hurry.
Examples of expressions that are easily included in hoaxes
Emphasized expression Very, quite, very, absolutely, all, surely, serious, etc.
Expressions that arouse anxiety Dangerous, malicious, difficult, etc.
Expression to hurry Immediately, urgently, etc.
Content related to life and money
Many life-threatening contents such as evacuation information and disaster information are also abused. In addition, money-related items such as "free services will be charged from next month" and "paid services will be free for a limited time" are also easily abused, so be sure to check the information source.
Source not listed
If it's true, but it doesn't have a link (URL of the source) or rationale, be sure to search for it yourself to find out the source. If it is related to government offices or companies, please check the official website or official blog.
It is also effective to check TV, radio, etc. depending on the content such as weather information.
Written in hearsay format
You should also be careful if the information provided includes hearsay formats such as "likely," "like," and "likely." If the source of information is ambiguous, such as "friends", "acquaintances", or "entertainers", please check with the parties directly or by searching the official website.
Recommended to spread
There are two main types of writing that recommend diffusion: malicious and good faith. In either case, it's important to know the source and not spread indiscriminately.
Examples of malicious spread demands
"If you don't give it to someone within ○ hours after seeing it, you will be unhappy."
"People who stop this information will be identified and attacked (captured)" etc.
An example of a request for spread in good faith
"The pet shop has collapsed, so please spread it to people who are likely to get it."
"Please share it with all your acquaintances to prevent damage" etc.
In other special cases, the source of information itself may be a false site (so-called joke site). Some people believe in content that is not common sense. If you receive such information, get in the habit of verifying the authenticity of the information.```
- 1
note : translated jp police advice HP above
2021-05-05
⚠️ *Action required - upgrade app for g0v* ⚠️
Per <https://g0v.hackmd.io/6ZwH7NANRqG6s-CCIy9IYw?both#yegogo-%E8%99%95%E7%90%86|20210428 discussion>, since we backup original reply text in the spreadsheet, we can consider removing spammer's reply text entirely. We can modify `removeArticleReply.js` so that it optionally take a CLI argument that replaces the reply text with specified text. We can provide link to the spreadsheet or takedown note in the replaced text.
Seems that there are missing kanji font in our current font selection <https://user-images.githubusercontent.com/108608/117100608-b8b55500-ada6-11eb-8b4e-8fe32bb6b6c0.png|image>
2021-05-07
⚠️ *Action required - upgrade app for g0v* ⚠️
Bumps <https://github.com/lodash/lodash|lodash> from 4.17.19 to 4.17.21. Commits • <https://github.com/lodash/lodash/commit/f299b52f39486275a9e6483b60a410e06520c538|`f299b52`> Bump to v4.17.21 • <https://github.com/lodash/lodash/commit/c4847ebe7d14540bb28a8b932a9ce1b9ecbfee1a|`c4847eb`> Improve performance of `toNumber`, `trim` and `trimEnd` on large input strings • <https://github.com/lodash/lodash/commit/3469357cff396a26c363f8c1b5a91dde28ba4b1c|`3469357`> Prevent command injection through `_.template`'s `variable` option • <https://github.com/lodash/lodash/commit/ded9bc66583ed0b4e3b7dc906206d40757b4a90a|`ded9bc6`> Bump to v4.17.20. • <https://github.com/lodash/lodash/commit/63150ef7645ac07961b63a86490f419f356429aa|`63150ef`> Documentation fixes. • <https://github.com/lodash/lodash/commit/00f0f62a979d2f5fa0287c06eae70cf9a62d8794|`00f0f62`> test.js: Remove trailing comma. • <https://github.com/lodash/lodash/commit/846e434c7a5b5692c55ebf5715ed677b70a32389|`846e434`> Temporarily use a custom fork of `lodash-cli`. • <https://github.com/lodash/lodash/commit/5d046f39cbd27f573914768e3b36eeefcc4f1229|`5d046f3`> Re-enable Travis tests on `4.17` branch. • <https://github.com/lodash/lodash/commit/aa816b36d402a1ad9385142ce7188f17dae514fd|`aa816b3`> Remove `/npm-package`. • See full diff in <https://github.com/lodash/lodash/compare/4.17.19...4.17.21|compare view> Maintainer changes This version was pushed to npm by <https://www.npmjs.com/~bnjmnt4n|bnjmnt4n>, a new releaser for lodash since your current version. <https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores|Dependabot compatibility score> Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. * * * Dependabot commands and options You can trigger Dependabot actions by commenting on this PR: • `@dependabot rebase` will rebase this PR • `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it • `@dependabot merge` will merge this PR after your CI passes on it • `@dependabot squash and merge` will squash and merge this PR after your CI passes on it • `@dependabot cancel merge` will cancel a previously requested merge and block automerging • `@dependabot reopen` will reopen this PR if it is closed • `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually • `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) • `@dependabot use these labels` will set the current labels as the default for future PRs for this repo and language • `@dependabot use these reviewers` will set the current reviewers as the default for future PRs for this repo and language • `@dependabot use these assignees` will set the current assignees as the default for future PRs for this repo and language • `@dependabot use this milestone` will set the current milestone as the default for future PRs for this repo and language You can disable automated security fix PRs for this repo from the <https://github.com/cofacts/rumors-site/network/alerts|Security Alerts page>.
Bumps <https://github.com/wycats/handlebars.js|handlebars> from 4.5.1 to 4.7.7. Changelog _Sourced from <https://github.com/handlebars-lang/handlebars.js/blob/master/release-notes.md|handlebars's changelog>._ > *v4.7.7 - February 15th, 2021* > > • fix weird error in integration tests - eb860c0 > • fix: check prototype property access in strict-mode (<https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1736|#1736>) - b6d3de7 > • fix: escape property names in compat mode (<https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1736|#1736>) - f058970 > • refactor: In spec tests, use expectTemplate over equals and shouldThrow (<https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1683|#1683>) - 77825f8 > • chore: start testing on Node.js 12 and 13 - 3789a30 > > (POSSIBLY) BREAKING CHANGES: > > • the changes from version <https://github.com/handlebars-lang/handlebars.js/blob/master/release-notes.md#v460---january-8th-2020|4.6.0> now also apply in when using the compile-option "strict: true". Access to prototype properties is forbidden completely by default, specific properties or methods can be allowed via runtime-options. See <https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1633|#1633> for details. If you are using Handlebars as documented, you should not be accessing prototype properties from your template anyway, so the changes should not be a problem for you. Only the use of undocumented features can break your build. > > That is why we only bump the patch version despite mentioning breaking changes. > > <https://github.com/wycats/handlebars.js/compare/v4.7.6...v4.7.7|Commits> > > *v4.7.6 - April 3rd, 2020* > > Chore/Housekeeping: > > • <https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1672|#1672> - Switch cmd parser to latest minimist (<https://api.github.com/users/dougwilson|`@dougwilson`> > > Compatibility notes: > > • Restored Node.js compatibility > > <https://github.com/wycats/handlebars.js/compare/v4.7.5...v4.7.6|Commits> > > *v4.7.5 - April 2nd, 2020* > > Chore/Housekeeping: > > • ~Node.js version support has been changed to v6+~ Reverted in 4.7.6 > > Compatibility notes: > > • ~Node.js < v6 is no longer supported~ Reverted in 4.7.6 > > <https://github.com/wycats/handlebars.js/compare/v4.7.4...v4.7.5|Commits> > > *v4.7.4 - April 1st, 2020* > > Chore/Housekeeping: > > • <https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1666|#1666> - Replaced minimist with yargs for handlebars CLI (<https://api.github.com/users/aorinevo|`@aorinevo`>, <https://api.github.com/users/AviVahl|`@AviVahl`> & <https://api.github.com/users/fabb|`@fabb`>) > > Compatibility notes: ... (truncated) Commits • <https://github.com/handlebars-lang/handlebars.js/commit/a9a8e403213583ca90cb7c872d3a22796c37d961|`a9a8e40`> v4.7.7 • <https://github.com/handlebars-lang/handlebars.js/commit/e66aed5b99c1b6c93564f37d627e34e5d60eb76e|`e66aed5`> Update release notes • <https://github.com/handlebars-lang/handlebars.js/commit/7d4d170ce46a53084a41920c5c7387c131357989|`7d4d170`> disable IE in Saucelabs tests • <https://github.com/handlebars-lang/handlebars.js/commit/eb860c08998f8f506360d305d89e1f4b40f72a0a|`eb860c0`> fix weird error in integration tests • <https://github.com/handlebars-lang/handlebars.js/commit/b6d3de7123eebba603e321f04afdbae608e8fea8|`b6d3de7`> fix: check prototype property access in strict-mode (<https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1736|#1736>) • <https://github.com/handlebars-lang/handlebars.js/commit/f0589701698268578199be25285b2ebea1c1e427|`f058970`> fix: escape property names in compat mode (<https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1736|#1736>) • <https://github.com/handlebars-lang/handlebars.js/commit/77825f8d3522356feb8e4160fac16344104d192b|`77825f8`> refator: In spec tests, use expectTemplate over equals and shouldThrow (<https://github-redirect.dependabot.com/wycats/handlebars.js/issues/1683|#1683>) • <https://github.com/handlebars-lang/handlebars.js/commit/3789a309554fd600caeae442f40881cf93eb3b54|`3789a30`> chore: start testing on Node.js 12 and 13 • <https://github.com/handlebars-lang/handlebars.js/commit/e6ad93ea01bcde1f8ddaa4b4ebe572dd616abfaa|`e6ad93e`> v4.7.6 • <https://github.com/handlebars-lang/handlebars.js/commit/2bf4fc6fd3ae3d8f076d628653f284d85faebeb4|`2bf4fc6`> Update release notes • Additional commits viewable in <https://github.com/wycats/handlebars.js/compare/v4.5.1...v4.7.7|compare view> <https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores|Dependabot compatibility score> Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. * * * Dependabot commands and options You can trigger Dependabot actions by commenting on this PR: • `@dependabot rebase` will rebase this PR • `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it • `@dependabot merge` will merge this PR after your CI passes on it • `@dependabot squash and merge` will squash and merge this PR after your CI passes on it • `@dependabot cancel merge` will cancel a previously requested merge and block automerging • `@dependabot reopen` will reopen this PR if it is closed • `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually • `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) • `@dependabot use these labels` will set the current labels as the default for future PRs for this repo and language • `@dependabot use these reviewers` will set the current reviewers as the default for future PRs for this repo and language • `@dependabot use these assignees` will set the current assignees as the default for future PRs for this repo and language • `@dependabot use this milestone` will set the current milestone as the default for future PRs for this repo and language You can disable automated security fix PRs for this repo from the <https://github.com/cofacts/rumors-fb-bot/network/alerts|Security Alerts page>.
Superseded by <https://github.com/cofacts/rumors-fb-bot/pull/22|#22>.
2021-05-08
Bumps <https://github.com/lodash/lodash|lodash> from 4.17.10 to 4.17.21. Commits • <https://github.com/lodash/lodash/commit/f299b52f39486275a9e6483b60a410e06520c538|`f299b52`> Bump to v4.17.21 • <https://github.com/lodash/lodash/commit/c4847ebe7d14540bb28a8b932a9ce1b9ecbfee1a|`c4847eb`> Improve performance of `toNumber`, `trim` and `trimEnd` on large input strings • <https://github.com/lodash/lodash/commit/3469357cff396a26c363f8c1b5a91dde28ba4b1c|`3469357`> Prevent command injection through `_.template`'s `variable` option • <https://github.com/lodash/lodash/commit/ded9bc66583ed0b4e3b7dc906206d40757b4a90a|`ded9bc6`> Bump to v4.17.20. • <https://github.com/lodash/lodash/commit/63150ef7645ac07961b63a86490f419f356429aa|`63150ef`> Documentation fixes. • <https://github.com/lodash/lodash/commit/00f0f62a979d2f5fa0287c06eae70cf9a62d8794|`00f0f62`> test.js: Remove trailing comma. • <https://github.com/lodash/lodash/commit/846e434c7a5b5692c55ebf5715ed677b70a32389|`846e434`> Temporarily use a custom fork of `lodash-cli`. • <https://github.com/lodash/lodash/commit/5d046f39cbd27f573914768e3b36eeefcc4f1229|`5d046f3`> Re-enable Travis tests on `4.17` branch. • <https://github.com/lodash/lodash/commit/aa816b36d402a1ad9385142ce7188f17dae514fd|`aa816b3`> Remove `/npm-package`. • <https://github.com/lodash/lodash/commit/d7fbc52ee0466a6d248f047b5d5c3e6d1e099056|`d7fbc52`> Bump to v4.17.19 • Additional commits viewable in <https://github.com/lodash/lodash/compare/4.17.10...4.17.21|compare view> Maintainer changes This version was pushed to npm by <https://www.npmjs.com/~bnjmnt4n|bnjmnt4n>, a new releaser for lodash since your current version. <https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores|Dependabot compatibility score> Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. * * * Dependabot commands and options You can trigger Dependabot actions by commenting on this PR: • `@dependabot rebase` will rebase this PR • `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it • `@dependabot merge` will merge this PR after your CI passes on it • `@dependabot squash and merge` will squash and merge this PR after your CI passes on it • `@dependabot cancel merge` will cancel a previously requested merge and block automerging • `@dependabot reopen` will reopen this PR if it is closed • `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually • `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) • `@dependabot use these labels` will set the current labels as the default for future PRs for this repo and language • `@dependabot use these reviewers` will set the current reviewers as the default for future PRs for this repo and language • `@dependabot use these assignees` will set the current assignees as the default for future PRs for this repo and language • `@dependabot use this milestone` will set the current milestone as the default for future PRs for this repo and language You can disable automated security fix PRs for this repo from the <https://github.com/cofacts/rumors-fb-bot/network/alerts|Security Alerts page>.
Superseded by <https://github.com/cofacts/rumors-fb-bot/pull/23|#23>.
@mrorz Hi! Sorry but John left our team this week. I will let you know our possible steps in meantime.
- 👌1
@mrorz BTW is your name Mr.オルツ like geek Twitter Japanese? Just wondered
Good quesiton, I haven’t thought of how the ID can be spelled in カタカナ lol
調整了一下 Article LIFF feedback form
• upvote / downvote 與 reason text input 會放一起填
• 送出後會進入 summary，可回到編輯模式或展開其他人的 feedback
◦ Summary 高度最矮，僅用刷淡底色標注使用者的選擇。這是希望使用者繼續往下捲到其他回應 or 下面的內容，以不干擾使用者繼續往下讀為主。
◦ 之前討論的「分頁」形式我覺得還是太複雜，會不知道點下去會是編輯還是看其他人回應還怎樣，而且會干擾使用者繼續往下讀。
• 其他人的 feedback 是 upvote / downvote 人混排
https://www.figma.com/file/DvmAQjMJCncuPORWKnljM1/Cofacts-website-MrOrz?node-id=3042%3A2
• upvote / downvote 與 reason text input 會放一起填
• 送出後會進入 summary，可回到編輯模式或展開其他人的 feedback
◦ Summary 高度最矮，僅用刷淡底色標注使用者的選擇。這是希望使用者繼續往下捲到其他回應 or 下面的內容，以不干擾使用者繼續往下讀為主。
◦ 之前討論的「分頁」形式我覺得還是太複雜，會不知道點下去會是編輯還是看其他人回應還怎樣，而且會干擾使用者繼續往下讀。
• 其他人的 feedback 是 upvote / downvote 人混排
https://www.figma.com/file/DvmAQjMJCncuPORWKnljM1/Cofacts-website-MrOrz?node-id=3042%3A2
Figma
Good quesiton, I haven’t thought of how the ID can be spelled in カタカナ lol
⚠️ *Action required - upgrade app for g0v* ⚠️
In reply page, the sidebar title should be "Similar replies" rather than "Similar messages". <https://user-images.githubusercontent.com/108608/117537599-96386b80-b034-11eb-8cdc-f3f0de947d41.png|image>
Discussions: <https://g0v.hackmd.io/BRQtYUVzTU2vNxainshnMQ#%E4%BD%BF%E7%94%A8%E8%80%85%E8%AE%8A%E5%A4%9A|https://g0v.hackmd.io/BRQtYUVzTU2vNxainshnMQ#%E4%BD%BF%E7%94%A8%E8%80%85%E8%AE%8A%E5%A4%9A> *Screenshots* *Message not in DB* <https://user-images.githubusercontent.com/108608/117537838-127f7e80-b036-11eb-923d-6990ebc6e375.png|image> *There are similar messages in DB but not what user want* <https://user-images.githubusercontent.com/108608/117537856-29be6c00-b036-11eb-826f-d1e080ea28a5.png|image> *Message in DB but not replied yet* <https://user-images.githubusercontent.com/108608/117537875-38a51e80-b036-11eb-94f6-158893f5c36f.png|image>
*Pull Request Test Coverage Report for <https://coveralls.io/builds/39483780|Build 1320>* • *0* of *0* changed or added relevant lines in *0* files are covered. • No unchanged relevant lines lost coverage. • Overall coverage remained the same at *86.629%* * * * * * * *:yellow_heart: - <https://coveralls.io|Coveralls>*
2021-05-09
TODO: • tesseract support • automatic build & push in Github Actions with `.env` as secret (<https://docs.docker.com/ci-cd/github-actions/|https://docs.docker.com/ci-cd/github-actions/>)
*Pull Request Test Coverage Report for <https://coveralls.io/builds/39487586|Build 1326>* • *0* of *0* changed or added relevant lines in *0* files are covered. • No unchanged relevant lines lost coverage. • Overall coverage remained the same at *86.629%* * * * * * * *:yellow_heart: - <https://coveralls.io|Coveralls>*
Bumps <https://github.com/npm/hosted-git-info|hosted-git-info> from 2.8.4 to 2.8.9. Changelog _Sourced from <https://github.com/npm/hosted-git-info/blob/v2.8.9/CHANGELOG.md|hosted-git-info's changelog>._ > *<https://github.com/npm/hosted-git-info/compare/v2.8.8...v2.8.9|2.8.9> (2021-04-07)* > *Bug Fixes* > > • backport regex fix from <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/76|#76> (<https://github.com/npm/hosted-git-info/commit/29adfe5|29adfe5>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/84|#84> > > *<https://github.com/npm/hosted-git-info/compare/v2.8.7...v2.8.8|2.8.8> (2020-02-29)* > *Bug Fixes* > > • <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/61|#61> & <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/65|#65> addressing issues w/ url.URL implmentation which regressed node 6 support (<https://github.com/npm/hosted-git-info/commit/5038b18|5038b18>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/66|#66> > > *<https://github.com/npm/hosted-git-info/compare/v2.8.6...v2.8.7|2.8.7> (2020-02-26)* > *Bug Fixes* > > • Do not attempt to use url.URL when unavailable (<https://github.com/npm/hosted-git-info/commit/2d0bb66|2d0bb66>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/61|#61> <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/62|#62> > • Do not pass scp-style URLs to the WhatWG url.URL (<https://github.com/npm/hosted-git-info/commit/f2cdfcf|f2cdfcf>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/60|#60> > > *<https://github.com/npm/hosted-git-info/compare/v2.8.5...v2.8.6|2.8.6> (2020-02-25)* > > *<https://github.com/npm/hosted-git-info/compare/v2.8.4...v2.8.5|2.8.5> (2019-10-07)* > *Bug Fixes* > > • updated pathmatch for gitlab (<https://github.com/npm/hosted-git-info/commit/e8325b5|e8325b5>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/51|#51> > • updated pathmatch for gitlab (<https://github.com/npm/hosted-git-info/commit/ffe056f|ffe056f>) Commits • <https://github.com/npm/hosted-git-info/commit/8d4b3697d79bcd89cdb36d1db165e3696c783a01|`8d4b369`> chore(release): 2.8.9 • <https://github.com/npm/hosted-git-info/commit/29adfe5ef789784c861b2cdeb15051ec2ba651a7|`29adfe5`> fix: backport regex fix from <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/76|#76> • <https://github.com/npm/hosted-git-info/commit/afeaefdd86ba9bb5044be3c1554a666d007cf19a|`afeaefd`> chore(release): 2.8.8 • <https://github.com/npm/hosted-git-info/commit/5038b1891a61ca3cd7453acbf85d7011fe0086bb|`5038b18`> fix: <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/61|#61> & <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/65|#65> addressing issues w/ url.URL implmentation which regressed nod... • <https://github.com/npm/hosted-git-info/commit/7440afa859162051c191e55d8ecfaf69a193b026|`7440afa`> chore(release): 2.8.7 • <https://github.com/npm/hosted-git-info/commit/2d0bb6615ecb8f9ef1019bc0737aab7f6449641f|`2d0bb66`> fix: Do not attempt to use url.URL when unavailable • <https://github.com/npm/hosted-git-info/commit/f2cdfcf33ad2bd3bd1acdba0326281089f53c5b1|`f2cdfcf`> fix: Do not pass scp-style URLs to the WhatWG url.URL • <https://github.com/npm/hosted-git-info/commit/e1b83df5d9cb1f8bb220352e20565560548d2292|`e1b83df`> chore(release): 2.8.6 • <https://github.com/npm/hosted-git-info/commit/ff259a6117c62df488e927820e30bec2f7ee453f|`ff259a6`> Ensure passwords in hosted Git URLs are correctly escaped • <https://github.com/npm/hosted-git-info/commit/624fd6f301dd5a1fd7ad1b333d6f8921a12ff98c|`624fd6f`> chore(release): 2.8.5 • Additional commits viewable in <https://github.com/npm/hosted-git-info/compare/v2.8.4...v2.8.9|compare view> Maintainer changes This version was pushed to npm by <https://www.npmjs.com/~nlf|nlf>, a new releaser for hosted-git-info since your current version. <https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores|Dependabot compatibility score> Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. * * * Dependabot commands and options You can trigger Dependabot actions by commenting on this PR: • `@dependabot rebase` will rebase this PR • `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it • `@dependabot merge` will merge this PR after your CI passes on it • `@dependabot squash and merge` will squash and merge this PR after your CI passes on it • `@dependabot cancel merge` will cancel a previously requested merge and block automerging • `@dependabot reopen` will reopen this PR if it is closed • `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually • `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) • `@dependabot use these labels` will set the current labels as the default for future PRs for this repo and language • `@dependabot use these reviewers` will set the current reviewers as the default for future PRs for this repo and language • `@dependabot use these assignees` will set the current assignees as the default for future PRs for this repo and language • `@dependabot use this milestone` will set the current milestone as the default for future PRs for this repo and language You can disable automated security fix PRs for this repo from the <https://github.com/cofacts/rumors-site/network/alerts|Security Alerts page>.
2021-05-10
⚠️ *Action required - upgrade app for g0v* ⚠️
Bumps <https://github.com/npm/hosted-git-info|hosted-git-info> from 2.7.1 to 2.8.9. Changelog _Sourced from <https://github.com/npm/hosted-git-info/blob/v2.8.9/CHANGELOG.md|hosted-git-info's changelog>._ > *<https://github.com/npm/hosted-git-info/compare/v2.8.8...v2.8.9|2.8.9> (2021-04-07)* > *Bug Fixes* > > • backport regex fix from <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/76|#76> (<https://github.com/npm/hosted-git-info/commit/29adfe5|29adfe5>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/84|#84> > > *<https://github.com/npm/hosted-git-info/compare/v2.8.7...v2.8.8|2.8.8> (2020-02-29)* > *Bug Fixes* > > • <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/61|#61> & <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/65|#65> addressing issues w/ url.URL implmentation which regressed node 6 support (<https://github.com/npm/hosted-git-info/commit/5038b18|5038b18>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/66|#66> > > *<https://github.com/npm/hosted-git-info/compare/v2.8.6...v2.8.7|2.8.7> (2020-02-26)* > *Bug Fixes* > > • Do not attempt to use url.URL when unavailable (<https://github.com/npm/hosted-git-info/commit/2d0bb66|2d0bb66>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/61|#61> <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/62|#62> > • Do not pass scp-style URLs to the WhatWG url.URL (<https://github.com/npm/hosted-git-info/commit/f2cdfcf|f2cdfcf>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/60|#60> > > *<https://github.com/npm/hosted-git-info/compare/v2.8.5...v2.8.6|2.8.6> (2020-02-25)* > > *<https://github.com/npm/hosted-git-info/compare/v2.8.4...v2.8.5|2.8.5> (2019-10-07)* > *Bug Fixes* > > • updated pathmatch for gitlab (<https://github.com/npm/hosted-git-info/commit/e8325b5|e8325b5>), closes <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/51|#51> > • updated pathmatch for gitlab (<https://github.com/npm/hosted-git-info/commit/ffe056f|ffe056f>) > > *<https://github.com/npm/hosted-git-info/compare/v2.8.3...v2.8.4|2.8.4> (2019-08-12)* ... (truncated) Commits • <https://github.com/npm/hosted-git-info/commit/8d4b3697d79bcd89cdb36d1db165e3696c783a01|`8d4b369`> chore(release): 2.8.9 • <https://github.com/npm/hosted-git-info/commit/29adfe5ef789784c861b2cdeb15051ec2ba651a7|`29adfe5`> fix: backport regex fix from <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/76|#76> • <https://github.com/npm/hosted-git-info/commit/afeaefdd86ba9bb5044be3c1554a666d007cf19a|`afeaefd`> chore(release): 2.8.8 • <https://github.com/npm/hosted-git-info/commit/5038b1891a61ca3cd7453acbf85d7011fe0086bb|`5038b18`> fix: <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/61|#61> & <https://github-redirect.dependabot.com/npm/hosted-git-info/issues/65|#65> addressing issues w/ url.URL implmentation which regressed nod... • <https://github.com/npm/hosted-git-info/commit/7440afa859162051c191e55d8ecfaf69a193b026|`7440afa`> chore(release): 2.8.7 • <https://github.com/npm/hosted-git-info/commit/2d0bb6615ecb8f9ef1019bc0737aab7f6449641f|`2d0bb66`> fix: Do not attempt to use url.URL when unavailable • <https://github.com/npm/hosted-git-info/commit/f2cdfcf33ad2bd3bd1acdba0326281089f53c5b1|`f2cdfcf`> fix: Do not pass scp-style URLs to the WhatWG url.URL • <https://github.com/npm/hosted-git-info/commit/e1b83df5d9cb1f8bb220352e20565560548d2292|`e1b83df`> chore(release): 2.8.6 • <https://github.com/npm/hosted-git-info/commit/ff259a6117c62df488e927820e30bec2f7ee453f|`ff259a6`> Ensure passwords in hosted Git URLs are correctly escaped • <https://github.com/npm/hosted-git-info/commit/624fd6f301dd5a1fd7ad1b333d6f8921a12ff98c|`624fd6f`> chore(release): 2.8.5 • Additional commits viewable in <https://github.com/npm/hosted-git-info/compare/v2.7.1...v2.8.9|compare view> Maintainer changes This version was pushed to npm by <https://www.npmjs.com/~nlf|nlf>, a new releaser for hosted-git-info since your current version. <https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores|Dependabot compatibility score> Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. * * * Dependabot commands and options You can trigger Dependabot actions by commenting on this PR: • `@dependabot rebase` will rebase this PR • `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it • `@dependabot merge` will merge this PR after your CI passes on it • `@dependabot squash and merge` will squash and merge this PR after your CI passes on it • `@dependabot cancel merge` will cancel a previously requested merge and block automerging • `@dependabot reopen` will reopen this PR if it is closed • `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually • `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) • `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) • `@dependabot use these labels` will set the current labels as the default for future PRs for this repo and language • `@dependabot use these reviewers` will set the current reviewers as the default for future PRs for this repo and language • `@dependabot use these assignees` will set the current assignees as the default for future PRs for this repo and language • `@dependabot use this milestone` will set the current milestone as the default for future PRs for this repo and language You can disable automated security fix PRs for this repo from the <https://github.com/cofacts/rumors-fb-bot/network/alerts|Security Alerts page>.
週末~因為 2077 破完了所以~試著把 rumors-line-bot 弄成 docker image，未來打算搬進跟 API、site、elasticsearch 同樣的一台主機。如果試跑起來沒問題的話，之後就能擺脫 Heroku 的 512MB / 1024MB RAM、每天抱怨 memory quota exceed、以及每個月約 45 鎂的使用費囉。
現在 Docker 版 LINE bot 已經放到 staging https://lin.ee/1QUzEX4nI 大家可以測測。
Related PR: https://github.com/cofacts/rumors-line-bot/pull/253
現在 Docker 版 LINE bot 已經放到 staging https://lin.ee/1QUzEX4nI 大家可以測測。
Related PR: https://github.com/cofacts/rumors-line-bot/pull/253
• Update dockerfile • Multi-stage build to make docker image smaller • apply `.env` file in build process (which is consistent to Heroku build environment) • npm scripts: Move build process to `heroku-postbuild` • so that it runs on heroku only, not in docker builds • heroku build should still work • .travis.yml updated accordingly • Categorize env vars into "Build time variables" and "Runtime secrets" • "Build time variables" are used by webpack and will be built into docker image • "Runtime secrets" are given when running docker container • Cofacts API introspection script makes sure `data/` folder exist so that it works in docker build • Local `data/` folder contains mongodb garbage so it is ignored in docker • Thus we need to build our own `data/` folder within docker • README documentation • `tmp_image_process` does not trigger pm2 watch since it watches only `/src` folder. Paragraph in README removed. TODO: ☑︎ tesseract support ☑︎ Deploy to staging - <https://lin.ee/1QUzEX4nI|https://lin.ee/1QUzEX4nI> ☐ automatic build & push in Github Actions with `.env` as secret (<https://docs.docker.com/ci-cd/github-actions/|https://docs.docker.com/ci-cd/github-actions/> , <https://stackoverflow.com/a/59482124|https://stackoverflow.com/a/59482124>) ☐ README for docker image build ☐ Log drain
- 💪1
現在還缺
• 自動 build image。因為 rumors-line-bot build-time 有太多 env var，build 的時候要給 env file，比其他都還複雜，過往在 docker hub 直接 build 的狀況可能行不通。可能會改成在可以更彈性設定 secret 與 build label 的 github action 來 build + push。
• 捕捉對話 log 過去是使用 heroku log drain，但換到 linode 之後可能要想其他方式。我覺得一種做法是用 vector ( fluentd alternative in Rust, https://vector.dev/ ) 把 docker output 無腦接到 S3 存起來。
• 自動 build image。因為 rumors-line-bot build-time 有太多 env var，build 的時候要給 env file，比其他都還複雜，過往在 docker hub 直接 build 的狀況可能行不通。可能會改成在可以更彈性設定 secret 與 build label 的 github action 來 build + push。
• 捕捉對話 log 過去是使用 heroku log drain，但換到 linode 之後可能要想其他方式。我覺得一種做法是用 vector ( fluentd alternative in Rust, https://vector.dev/ ) 把 docker output 無腦接到 S3 存起來。
現在還缺
• 自動 build image。因為 rumors-line-bot build-time 有太多 env var，build 的時候要給 env file，比其他都還複雜，過往在 docker hub 直接 build 的狀況可能行不通。可能會改成在可以更彈性設定 secret 與 build label 的 github action 來 build + push。
• 捕捉對話 log 過去是使用 heroku log drain，但換到 linode 之後可能要想其他方式。我覺得一種做法是用 vector ( fluentd alternative in Rust, https://vector.dev/ ) 把 docker output 無腦接到 S3 存起來。
• 自動 build image。因為 rumors-line-bot build-time 有太多 env var，build 的時候要給 env file，比其他都還複雜，過往在 docker hub 直接 build 的狀況可能行不通。可能會改成在可以更彈性設定 secret 與 build label 的 github action 來 build + push。
• 捕捉對話 log 過去是使用 heroku log drain，但換到 linode 之後可能要想其他方式。我覺得一種做法是用 vector ( fluentd alternative in Rust, https://vector.dev/ ) 把 docker output 無腦接到 S3 存起來。
2021-05-11
2077好好玩judy很可愛
⚠️ *Action required - upgrade app for g0v* ⚠️
2021-05-12
2021-05-13
cofacts 最近有沒有被謠言回報轟炸 q_q
有
還有更棘手的，看起來像是從各級政府單位 leak 出來的、沒有證實過的確診者足跡。
還有更棘手的，看起來像是從各級政府單位 leak 出來的、沒有證實過的確診者足跡。
大家有什麼想法與經驗也歡迎分享唷
https://hackmd.io/@cofacts/r1EUpWTdu
https://hackmd.io/@cofacts/r1EUpWTdu
Replied to a thread: 2021-05-13 09:48:38
有
還有更棘手的，看起來像是從各級政府單位 leak 出來的、沒有證實過的確診者足跡。
還有更棘手的，看起來像是從各級政府單位 leak 出來的、沒有證實過的確診者足跡。
哦不 orz
oh no
`不要學我說話。` cofacts 好好笑哦
- 😆2
XDDD
❤️
`確診訊息太極鍵譜`
2
「真的假的」實在太強了！
求團長大力推廣（？）
我剛有在家裡群組推廣！
2
`如果看到訊息，不知道是不是真的，也可以傳到「Cofacts 真的假的」這個 LINE 帳號，他們是一群幫忙查證訊息內容的志工，如果是已經有人查證的內容，他們就會把結果回傳，可以多一點資訊參考 :)`
- 👍2
我現在是工人真的假的，先讓我爸傳訊息給我，我幫他傳給真的假的，我再回傳給他
- ❤️2
- 🦒1
\ @chiehg0v / ++
2021-05-14
⚠️ *Action required - upgrade app for g0v* ⚠️
Another instance is reported today by Robin, MyGoPen ``` 刷樂，依必朗，3M漱口水有含那成份Chlorhexidine <https://www.worldjournal.com/wj/story/121471/5394313?fbclid=IwAR1HowY-Wq-5szf51JMVqrkN5tM053M9-qCtEEjj-bSiIXT2JBGH7H9Ea> ``` <https://user-images.githubusercontent.com/108608/118162903-e4de6f00-b453-11eb-974c-b9a473b2ebaf.png|image> <https://user-images.githubusercontent.com/108608/118163066-1e16df00-b454-11eb-9c8f-e239744c91f6.png|image> Its variant will not trigger such error. No fbclid --> OK <https://user-images.githubusercontent.com/108608/118162947-f0319a80-b453-11eb-83c0-56e32ff67f35.png|image> URL only with fbclid --> OK <https://user-images.githubusercontent.com/108608/118162983-fb84c600-b453-11eb-934e-f1538aaa7407.png|image> Text only ---> OK <https://user-images.githubusercontent.com/108608/118163007-05a6c480-b454-11eb-853e-d117c18c6a39.png|image>
2021-05-15
covid19 疫情資訊野生整合平台有短網址啦 https://g0v.hackmd.io/@yitzu/covid-19/ 歡迎協作貢獻！
HackMD
本平台為 g0v 貢獻者彙整而成，非政府官方所設，所有內容僅供參考。轉傳資訊前，請注意資訊來源，勿協助傳遞不實資訊。
- 3
2
覺得打開jitsi 就可以跟大家一起工作好溫馨喔…….
- 🙌9
- 1
阿三好可愛
- 🙌6
Replied to a thread: 2021-05-13 09:48:38
大家有什麼想法與經驗也歡迎分享唷
https://hackmd.io/@cofacts/r1EUpWTdu
https://hackmd.io/@cofacts/r1EUpWTdu
# 確診足跡太極鍵譜 2021 年 5 月以來，網傳確診者足跡訊息大爆發。然而，我們也知道這些網傳的足跡常常未經證實，也不符合「[個案活動地點公布考慮原則](#「疫調流程」與「個案活動地點公布考
- 🙌3
@erickuo5124 has joined the channel
想詢問加入事實查核的連結
- 👍1
sns要ㄥˋ
用
找到ㄌ
⚠️ *Action required - upgrade app for g0v* ⚠️
*As-is* Currently the landing page only shows messages that is not replied yet. This may give new comer an initial impression that all messages are not getting replied in this website. <https://user-images.githubusercontent.com/108608/118363137-80dbb800-b5c5-11eb-866a-b934180bbd04.png|image> *The fix* Include messages that is being replied in landing page. <https://user-images.githubusercontent.com/108608/118363184-c304f980-b5c5-11eb-9789-2c085617cc96.png|image>
是說教學頁面的連結
好像可以放在多一點地方
例如說 article page 闢謠的部分
或者是「等你來答」上面多一條 info 讓使用者自己關掉之類
好像可以放在多一點地方
例如說 article page 闢謠的部分
或者是「等你來答」上面多一條 info 讓使用者自己關掉之類
https://cofacts.g0v.tw/tutorial?tab=bust-hoaxes
的確可以放在更多讓新手友善的地方，比如說各列表頁上面多加一個 banner
的確可以放在更多讓新手友善的地方，比如說各列表頁上面多加一個 banner
2021-05-16
@guo-jim has joined the channel
@karmin.sphinx has joined the channel
@joy has joined the channel
@vivilittlebao has joined the channel
指揮官：政府有責任設立一個地方，讓民眾可以確認疫情相關訊息
@kentwelcome has joined the channel
陳副：昨天就有收到，有人謊報確診，詐騙個人資料
- 😯1
@rubyful has joined the channel
@bettyvisualtech has joined the channel
@death1048576 如果 ESRI 轉經緯度的經緯度放在 Google My Maps 會歪歪的話，那會不會拿來跟 google timeline 的資料比對的時候也歪歪的呀囧
我們目前是抓 100m 的範圍，應該還好吧(？)
不過 Google Map 標的地點（店家本體）有時候就會和地址標出來的不一樣。只能盡力做吧
- 1
https://cofacts.g0v.tw/tutorial?tab=bust-hoaxes
的確可以放在更多讓新手友善的地方，比如說各列表頁上面多加一個 banner
的確可以放在更多讓新手友善的地方，比如說各列表頁上面多加一個 banner
@shouldwang has joined the channel
2021-05-17
@s09454 has joined the channel
@sfyou614 has joined the channel
@juancamilosc has joined the channel
2021-05-18
⚠️ *Action required - upgrade app for g0v* ⚠️
According to <https://developers.line.biz/en/reference/messaging-api/#send-multicast-message|API documentation>, we cannot send more than 500 users at once when multicasting. However, recently we are sending update messages to more than 500 users at once. Related Rollbar error: <https://rollbar.com/mrorz/rumors-line-bot/items/376/occurrences/164470118640/|https://rollbar.com/mrorz/rumors-line-bot/items/376/occurrences/164470118640/> We should batch these requests and re-send the notifications.
明天開會我們改線上唷
想要用 Tico 還是用 gather.town 呢～？
想要用 Tico 還是用 gather.town 呢～？
Gather town cofacts 小鎮
2021-05-19
gather.town
Gather is a video-calling space that lets multiple people hold separate conversations in parallel, walking in and out of those conversations just as easily as they would in real life.
*Pull Request Test Coverage Report for <https://coveralls.io/builds/39790390|Build 1338>* • *8* of *8* *(100.0%)* changed or added relevant lines in *1* file are covered. • No unchanged relevant lines lost coverage. • Overall coverage increased (+*0.4%*) to *87.005%* * * * * * * *:yellow_heart: - <https://coveralls.io|Coveralls>*
@takayuki.yes has joined the channel
⚠️ *Action required - upgrade app for g0v* ⚠️
今天晚上需要先去看個醫生，10:00 再開會唷
@arrrmac has joined the channel
Replied to a thread: 2021-05-19 14:34:19
今天晚上需要先去看個醫生，10:00 再開會唷
@may3104 has joined the channel
@alisaxdgogogo has joined the channel
@lucia761126 has joined the channel
2021-05-20
@ytsai1 has joined the channel
@guo-jim has left the channel
@b01601029 has joined the channel
@lue41517 has joined the channel
@guo-jim has joined the channel
@ducksteven has joined the channel
@ttps860823 has joined the channel
@cooshif has joined the channel
@denkenie has joined the channel
@s12i.tw has joined the channel
@adaam has joined the channel
大家好，我叫昌豪，是一名在北部醫院工作的住院醫師，
疫情近日急遽升溫，近20名來自全台各個醫院的住院醫師和我組成「實證醫學救台灣」的團隊，合力翻譯國際最新的醫療文獻與內容，並提供給網路媒體撰寫成新聞稿發佈。
目前，我們目前面臨的一個困難是，網路上有很多已經存在的資訊，但真偽不明且資料來源難以查找，我們很希望可以在g0v開啟一個「醫療闢謠專案」，是彙整、蒐集在網路上散播的中文、與疫情有關連的醫療資訊。
一方面我們希望翻譯的東西儘可能不要跟這些資訊重複，二方面其中的資訊有誤也可以給我們的醫師檢覈。最後，經過查驗是正確的資訊也可以直接開源給所有民眾取得。
目前Cofacts上的資訊種類相當多元，期望之後可以將需要醫療專業判讀的資訊獨立出來，並整理出一個專區。
另外，我們團隊近日有持續在翻譯國際文獻，也想確認之後是否可以將翻譯出的文獻在共筆上另外建立一個區域開源
https://g0v.hackmd.io/Iu8dI9H9TcePIL0Xl2xM4Q
由於小弟是第一次加進g0v的slack，對於運作的形式還不是很熟悉，希望能請各位前輩多多海涵、指教，謝謝大家
疫情近日急遽升溫，近20名來自全台各個醫院的住院醫師和我組成「實證醫學救台灣」的團隊，合力翻譯國際最新的醫療文獻與內容，並提供給網路媒體撰寫成新聞稿發佈。
目前，我們目前面臨的一個困難是，網路上有很多已經存在的資訊，但真偽不明且資料來源難以查找，我們很希望可以在g0v開啟一個「醫療闢謠專案」，是彙整、蒐集在網路上散播的中文、與疫情有關連的醫療資訊。
一方面我們希望翻譯的東西儘可能不要跟這些資訊重複，二方面其中的資訊有誤也可以給我們的醫師檢覈。最後，經過查驗是正確的資訊也可以直接開源給所有民眾取得。
目前Cofacts上的資訊種類相當多元，期望之後可以將需要醫療專業判讀的資訊獨立出來，並整理出一個專區。
另外，我們團隊近日有持續在翻譯國際文獻，也想確認之後是否可以將翻譯出的文獻在共筆上另外建立一個區域開源
https://g0v.hackmd.io/Iu8dI9H9TcePIL0Xl2xM4Q
由於小弟是第一次加進g0v的slack，對於運作的形式還不是很熟悉，希望能請各位前輩多多海涵、指教，謝謝大家
9- 2
太棒了，當然可以。歡迎直接共筆。右上角有編輯鍵。
目前 Cofacts 裡「疾病、醫藥」分類、「COVID-19 疫情」分類都是手工標注的，還沒有
而關於 @lue41517 這個部分
> 目前Cofacts上的資訊種類相當多元，期望之後可以將需要醫療專業判讀的資訊獨立出來，並整理出一個專區。
我有一個想法：
1. 「實證醫學救台灣」團隊可以自己架部落格（blogger, wordpress, medium, matters 都可以）來擺放自己想要放的任何文章
2. 謠言方面，來 Cofacts 看訊息的熱門程度；如果在 LINE 遇到醫療謠言，就傳到 Cofacts 累積熱門程度。
3. 如果想要選謠言來寫文章的話，可以來 Cofacts 自由挑選，寫完之後發表在自己的部落格
4. 把寫好的部落格文章，在 Cofacts 上發表為該謠言的闢謠回應。這樣一來， Cofacts 的使用者以及其下游（如美玉姨）的使用者在查該謠言的時候，就會看到你們的文章、連進你們的貼文裡頭，帶來閱覽量。
這個把 Cofacts 作為謠言來源、以及導流管道的使用模式，與 MyGoPen、TFC、蘭姆酒吐司等事實查核單位用 Cofacts 的方式相同，可以完全確保運作獨立，又讓資料與流量互通有無～
> 目前Cofacts上的資訊種類相當多元，期望之後可以將需要醫療專業判讀的資訊獨立出來，並整理出一個專區。
我有一個想法：
1. 「實證醫學救台灣」團隊可以自己架部落格（blogger, wordpress, medium, matters 都可以）來擺放自己想要放的任何文章
2. 謠言方面，來 Cofacts 看訊息的熱門程度；如果在 LINE 遇到醫療謠言，就傳到 Cofacts 累積熱門程度。
3. 如果想要選謠言來寫文章的話，可以來 Cofacts 自由挑選，寫完之後發表在自己的部落格
4. 把寫好的部落格文章，在 Cofacts 上發表為該謠言的闢謠回應。這樣一來， Cofacts 的使用者以及其下游（如美玉姨）的使用者在查該謠言的時候，就會看到你們的文章、連進你們的貼文裡頭，帶來閱覽量。
這個把 Cofacts 作為謠言來源、以及導流管道的使用模式，與 MyGoPen、TFC、蘭姆酒吐司等事實查核單位用 Cofacts 的方式相同，可以完全確保運作獨立，又讓資料與流量互通有無～
Cofacts 另外還有公開每則謠言的流量數據在 http://cofacts.tw/analytics
從這裡選題，應該對於選到熱門訊息有幫助唷！
從這裡選題，應該對於選到熱門訊息有幫助唷！
了解了，我會跟團隊討論看看怎麼做比較合適，謝謝你們
感謝感謝
Cofacts 這裏也很希望這些醫藥相關的資訊，可以有專業人士的回應
非常感激醫師們發起這個團隊 m(_ _)m
Cofacts 這裏也很希望這些醫藥相關的資訊，可以有專業人士的回應
非常感激醫師們發起這個團隊 m(_ _)m
另外想請教，若訊息來源不是文字，而是圖片或影片，也可以回報cofacts嗎
如果圖片或影片是網址的話沒有問題
圖片本身的話，可能需要手打裡面的字送進資料庫，LINE bot 未來收到同一張圖的時候，（有機會）能夠透過 Tesseract OCR 辨識出一樣的字進而回傳
影片的話就無法了 QQ
圖片本身的話，可能需要手打裡面的字送進資料庫，LINE bot 未來收到同一張圖的時候，（有機會）能夠透過 Tesseract OCR 辨識出一樣的字進而回傳
影片的話就無法了 QQ
恩好
因為滿多LINE上面的假消息都是以圖片的型態散播
因為滿多LINE上面的假消息都是以圖片的型態散播
如果是youtube 影片的話都會收到回應喔，以圖片散佈的假訊息也沒有問題
@cyberbsd has joined the channel
@yiwenwang0713 has joined the channel
太棒了，當然可以。歡迎直接共筆。右上角有編輯鍵。
目前 Cofacts 裡「疾病、醫藥」分類、「COVID-19 疫情」分類都是手工標注的，還沒有
而關於 @lue41517 這個部分
> 目前Cofacts上的資訊種類相當多元，期望之後可以將需要醫療專業判讀的資訊獨立出來，並整理出一個專區。
我有一個想法：
1. 「實證醫學救台灣」團隊可以自己架部落格（blogger, wordpress, medium, matters 都可以）來擺放自己想要放的任何文章
2. 謠言方面，來 Cofacts 看訊息的熱門程度；如果在 LINE 遇到醫療謠言，就傳到 Cofacts 累積熱門程度。
3. 如果想要選謠言來寫文章的話，可以來 Cofacts 自由挑選，寫完之後發表在自己的部落格
4. 把寫好的部落格文章，在 Cofacts 上發表為該謠言的闢謠回應。這樣一來， Cofacts 的使用者以及其下游（如美玉姨）的使用者在查該謠言的時候，就會看到你們的文章、連進你們的貼文裡頭，帶來閱覽量。
這個把 Cofacts 作為謠言來源、以及導流管道的使用模式，與 MyGoPen、TFC、蘭姆酒吐司等事實查核單位用 Cofacts 的方式相同，可以完全確保運作獨立，又讓資料與流量互通有無～
> 目前Cofacts上的資訊種類相當多元，期望之後可以將需要醫療專業判讀的資訊獨立出來，並整理出一個專區。
我有一個想法：
1. 「實證醫學救台灣」團隊可以自己架部落格（blogger, wordpress, medium, matters 都可以）來擺放自己想要放的任何文章
2. 謠言方面，來 Cofacts 看訊息的熱門程度；如果在 LINE 遇到醫療謠言，就傳到 Cofacts 累積熱門程度。
3. 如果想要選謠言來寫文章的話，可以來 Cofacts 自由挑選，寫完之後發表在自己的部落格
4. 把寫好的部落格文章，在 Cofacts 上發表為該謠言的闢謠回應。這樣一來， Cofacts 的使用者以及其下游（如美玉姨）的使用者在查該謠言的時候，就會看到你們的文章、連進你們的貼文裡頭，帶來閱覽量。
這個把 Cofacts 作為謠言來源、以及導流管道的使用模式，與 MyGoPen、TFC、蘭姆酒吐司等事實查核單位用 Cofacts 的方式相同，可以完全確保運作獨立，又讓資料與流量互通有無～
Cofacts 另外還有公開每則謠言的流量數據在 http://cofacts.tw/analytics
從這裡選題，應該對於選到熱門訊息有幫助唷！
從這裡選題，應該對於選到熱門訊息有幫助唷！
了解了，我會跟團隊討論看看怎麼做比較合適，謝謝你們
感謝感謝
Cofacts 這裏也很希望這些醫藥相關的資訊，可以有專業人士的回應
非常感激醫師們發起這個團隊 m(_ _)m
Cofacts 這裏也很希望這些醫藥相關的資訊，可以有專業人士的回應
非常感激醫師們發起這個團隊 m(_ _)m
⚠️ *Action required - upgrade app for g0v* ⚠️
1. Comment out logging `JSON.stringify(result)` as it will crash the terminal when there are 10000 users to send 2. Send incremental `console.log`s so that we can track the processing status
*Pull Request Test Coverage Report for <https://coveralls.io/builds/39843825|Build 1342>* • *2* of *2* *(100.0%)* changed or added relevant lines in *2* files are covered. • No unchanged relevant lines lost coverage. • Overall coverage increased (+*0.008%*) to *87.013%* * * * * * * *:yellow_heart: - <https://coveralls.io|Coveralls>*
另外想請教，若訊息來源不是文字，而是圖片或影片，也可以回報cofacts嗎
如果圖片或影片是網址的話沒有問題
圖片本身的話，可能需要手打裡面的字送進資料庫，LINE bot 未來收到同一張圖的時候，（有機會）能夠透過 Tesseract OCR 辨識出一樣的字進而回傳
影片的話就無法了 QQ
圖片本身的話，可能需要手打裡面的字送進資料庫，LINE bot 未來收到同一張圖的時候，（有機會）能夠透過 Tesseract OCR 辨識出一樣的字進而回傳
影片的話就無法了 QQ
@h6020812 has joined the channel
恩好
因為滿多LINE上面的假消息都是以圖片的型態散播
因為滿多LINE上面的假消息都是以圖片的型態散播
@appbestellt has joined the channel
2021-05-21
@ita3051 has joined the channel
Cofacts chatbot 有個功能，是會把新的回應主動推送給看過這則訊息的人。
5/15~5/19 這個訊息有壞掉一下，不過昨天修好之後，補推通知給 5,000 人，今日中午再推 9,800 人，告訴他們可以回 Cofacts 打開「看過的訊息」列表來查看新回應。
透過推播打開「看過的訊息」列表的人數，昨天與今天（到下午為止）各自都有 2,500 人。
Every reply counts!
5/15~5/19 這個訊息有壞掉一下，不過昨天修好之後，補推通知給 5,000 人，今日中午再推 9,800 人，告訴他們可以回 Cofacts 打開「看過的訊息」列表來查看新回應。
透過推播打開「看過的訊息」列表的人數，昨天與今天（到下午為止）各自都有 2,500 人。
Every reply counts!
- 🙌6
好棒！
好棒！
@vivi5377 has joined the channel
如果是youtube 影片的話都會收到回應喔，以圖片散佈的假訊息也沒有問題
打擾大家了，我是昌豪，一名住院醫師，
前幾日有上來和大家請教是否可以另外整理出一個醫療闢謠專案，與@mrorz和@bil 幾經討論之後，我們決定另外開一個臉書社團，直接將我們整理好的正確資訊發布在上面，未來可以直接供cofacts當作ref
也在社團當中增加一個專區讓民眾把真偽不明的訊息和網路上正確的疫情資訊整理上去，也可以把詢問度高的訊息丟到cofacts上累積流量
這是社團連結 https://bit.ly/3ft3Mi0
歡迎g0v的前輩們進社團裡面，不吝給予更正、指教，以及提供訊息，謝謝你們!
前幾日有上來和大家請教是否可以另外整理出一個醫療闢謠專案，與@mrorz和@bil 幾經討論之後，我們決定另外開一個臉書社團，直接將我們整理好的正確資訊發布在上面，未來可以直接供cofacts當作ref
也在社團當中增加一個專區讓民眾把真偽不明的訊息和網路上正確的疫情資訊整理上去，也可以把詢問度高的訊息丟到cofacts上累積流量
這是社團連結 https://bit.ly/3ft3Mi0
歡迎g0v的前輩們進社團裡面，不吝給予更正、指教，以及提供訊息，謝謝你們!
Facebook Groups
本社團主要宗旨為：傳遞正確的疫情消息 由民眾提問 - 醫師解惑 - 將資訊彙整。 「實證醫學救台灣」來自台灣各地的住院醫師組成的團隊，在疫情期間持續翻譯國際防疫相關文獻，協助解答疫情迷思，幫助台灣挺過疫情。 :speaking_head_in_silhouette: 民眾提問 歡迎民眾提出任何關於疫情相關的疑問 :female-doctor::skin-tone-2:醫師解惑 我們將邀請有專業醫療背景的醫生解惑 :bookmark_tabs:資訊彙整 彙整現有的疫情相關資訊，持續傳遞正確消息.
- ❤️13
- 🐧8
- 👍5
- 🦒3
謝謝你們
請問我可以把這個 group 分享到 facebook.com/groups/cofacts 邀請群組裡面的醫師一起加入嗎～
@lue41517 哇哇哇 好讚 \o/ ++
有點擔心 facebook group post 有 archive 不易的問題⋯但先運作看看，方便優先！（？）
我覺得累積一陣子之後再丟到部落格平台之類的也行
印尼群眾協作闢謠組織 Mafindo 也是很多 FB group 起家～
後來才架設資料庫網頁請志工把 FB 闢謠放進資料庫
是，我們希望在疫情初期以能夠更敏捷地反應謠言與民眾問題的社群媒體做開端，未來可以再慢慢整理上面的內容
謝謝@chihao 的醒
提醒
想確定一下
如果我們在 cofacts 上面看到醫學訊息不知道怎麼回的話，可以在這個 post reply 然後求救這樣嗎
https://www.facebook.com/groups/2560340054261472/permalink/2560442394251238/
如果我們在 cofacts 上面看到醫學訊息不知道怎麼回的話，可以在這個 post reply 然後求救這樣嗎
https://www.facebook.com/groups/2560340054261472/permalink/2560442394251238/
當然可以
我們也期待有足夠多的假消息，之後再正式用這些消息邀請更多醫師直接加入社團回復
@lue41517 & 醫師們 & cofacts ++ \o/
@howarder3 has joined the channel
我覺得累積一陣子之後再丟到部落格平台之類的也行
印尼群眾協作闢謠組織 Mafindo 也是很多 FB group 起家～
後來才架設資料庫網頁請志工把 FB 闢謠放進資料庫
是，我們希望在疫情初期以能夠更敏捷地反應謠言與民眾問題的社群媒體做開端，未來可以再慢慢整理上面的內容
謝謝@chihao 的醒
提醒
想確定一下
如果我們在 cofacts 上面看到醫學訊息不知道怎麼回的話，可以在這個 post reply 然後求救這樣嗎
https://www.facebook.com/groups/2560340054261472/permalink/2560442394251238/
如果我們在 cofacts 上面看到醫學訊息不知道怎麼回的話，可以在這個 post reply 然後求救這樣嗎
https://www.facebook.com/groups/2560340054261472/permalink/2560442394251238/
當然可以
我們也期待有足夠多的假消息，之後再正式用這些消息邀請更多醫師直接加入社團回復
2021-05-22
@alwaysgagaga has joined the channel
@kanelin21 has joined the channel
大家好～我是新加入！
- 🙌5
@lue41517 & 醫師們 & cofacts ++ \o/
@patata203302 has joined the channel
@philippe.chiu has joined the channel
當然可以!
社團是公開的，歡迎分享到任何地方
社團是公開的，歡迎分享到任何地方
有愈多人加入育好
愈好
2021-05-23
@tedisted has joined the channel
@yveschenyac has joined the channel
2021-05-24
⚠️ *Action required - upgrade app for g0v* ⚠️
Change deploy script to reflect <https://github.com/cofacts/rumors-line-bot/pull/253|cofacts/rumors-line-bot#253>
一大早就蔡壁如錄音之亂
蔡壁如：「我的聲音應該比較年輕啊⋯」
蔡壁如：「我的聲音應該比較年輕啊⋯」
https://cofacts.tw/article/1xhgdw5tchwom
https://cofacts.tw/article/2o277uut7h6qt
https://cofacts.tw/article/2wy9y6nx1jtcw
https://cofacts.tw/article/t1ezs2rdnebt
https://cofacts.tw/article/qpbicy794sjn
https://cofacts.tw/article/2elcd399skzoc
https://cofacts.tw/article/docdwuy7xj6j
https://cofacts.tw/article/1nstvh6fyec1m
https://cofacts.tw/article/n3labq7s8fd0
https://cofacts.tw/article/3c99tfa1sm7s3
https://cofacts.tw/article/2o277uut7h6qt
https://cofacts.tw/article/2wy9y6nx1jtcw
https://cofacts.tw/article/t1ezs2rdnebt
https://cofacts.tw/article/qpbicy794sjn
https://cofacts.tw/article/2elcd399skzoc
https://cofacts.tw/article/docdwuy7xj6j
https://cofacts.tw/article/1nstvh6fyec1m
https://cofacts.tw/article/n3labq7s8fd0
https://cofacts.tw/article/3c99tfa1sm7s3
XDD
這次是聲線不符被看穿，下次就難說了。現在的技術可以用別人的錄音剪輯合成一段新的錄音不是嗎 🥲
deep fake 什麼的等出現了再來煩惱吧
don’t get distracted from replying rumors
Replied to a thread: 2021-05-24 12:36:56
https://cofacts.tw/article/1xhgdw5tchwom
https://cofacts.tw/article/2o277uut7h6qt
https://cofacts.tw/article/2wy9y6nx1jtcw
https://cofacts.tw/article/t1ezs2rdnebt
https://cofacts.tw/article/qpbicy794sjn
https://cofacts.tw/article/2elcd399skzoc
https://cofacts.tw/article/docdwuy7xj6j
https://cofacts.tw/article/1nstvh6fyec1m
https://cofacts.tw/article/n3labq7s8fd0
https://cofacts.tw/article/3c99tfa1sm7s3
https://cofacts.tw/article/2o277uut7h6qt
https://cofacts.tw/article/2wy9y6nx1jtcw
https://cofacts.tw/article/t1ezs2rdnebt
https://cofacts.tw/article/qpbicy794sjn
https://cofacts.tw/article/2elcd399skzoc
https://cofacts.tw/article/docdwuy7xj6j
https://cofacts.tw/article/1nstvh6fyec1m
https://cofacts.tw/article/n3labq7s8fd0
https://cofacts.tw/article/3c99tfa1sm7s3
- 😮4
XDD
這次是聲線不符被看穿，下次就難說了。現在的技術可以用別人的錄音剪輯合成一段新的錄音不是嗎 🥲
關於 LINE bot 轉移到 Linode 的進度
請大家看一下這個 rumors-deploy 的 PR
之後預計會把 redis 與 LINE bot 搬到 Linode 上來作業
包含 LINE bot chat log 在內的 docker image stdout 會直接透過 docker engine 的 AWS log driver 送到 AWS cloudwatch（懶得用 vector 啦）
預計大小方面，五月開始到現在的對話紀錄大小為 1GB。Cloudwatch 有 5GB 的 free tier，大概五個月之後會開始付錢 (?)
請大家看一下這個 rumors-deploy 的 PR
之後預計會把 redis 與 LINE bot 搬到 Linode 上來作業
包含 LINE bot chat log 在內的 docker image stdout 會直接透過 docker engine 的 AWS log driver 送到 AWS cloudwatch（懶得用 vector 啦）
預計大小方面，五月開始到現在的對話紀錄大小為 1GB。Cloudwatch 有 5GB 的 free tier，大概五個月之後會開始付錢 (?)
@mrorz linode k8s?
cool 原來 linode 也有 k8s
雖然現在我們還是 docker-compose 打天下
我已經用很久了，@mrorz 搭配 github package 超好用
@mrorz 需要我幫你換 k8s 嗎
Linode k8s 的好處是 master 是不用錢的，只要付下面 node 費用😂😂
建議開三台，但是我省錢只開一台😂😂
聽起來很讚
所以是 rumors-deploy https://github.com/cofacts/rumors-deploy 的 docker-compose.yml 會變成 k8s 的 config 這樣嗎～～
所以是 rumors-deploy https://github.com/cofacts/rumors-deploy 的 docker-compose.yml 會變成 k8s 的 config 這樣嗎～～
看要不要直上 helm chart
他就會變 helm chart yaml
我只要管付錢跟開多大，剩下的 k8s 直接算這樣嗎
cooool
Db 的話 他會幫你開 linode storage
Mysql 會開 pod ，然後 pvc 會去掛 linode storage
我們是 elasticsearch
他有點肥
他有點肥
@mrorz 實際上就是你原本的 linode 花費
喔如果一台的話這樣嗎
只是想成不用自己進去管機器
變成管 k8s
那 cronjob 那些也是塞進 k8s ㄇ
我們現在還是有對 file system 做些髒髒事情
我們現在還是有對 file system 做些髒髒事情
例如說 bind mount 之類
這個要看你做了哪些😂
Cronjob的話 k8s 就 job scheduler
是說如果只有一台機器的話，那跟 docker-compose 差在哪裡 XD
雖然我是滿想拆小機器的
變成三台 20USD 的機器
變成三台 20USD 的機器
你可以開3台 20usd
但拆半天也就是省 20USD
因為我現在是一台 80USD
給 linode k8s 啊
對我來說 用 k8s 要加新服務或搬去其他 k8s 都很快啦
不過我個人目前是用 ko 做 deploy (所以一定要寫 golang lol)
感覺比較合適的做法會是
1. 先繼續把 rumors-line-bot 開在 linode 的計畫（畢竟先 dockerize 跑起來再說）
2. rumors-deploy 開一個 branch 把 docker-compose 改 helm chart
3. 試著把 staging 改成用 k8s deploy
4. 再想辦法如何弄 production
1. 先繼續把 rumors-line-bot 開在 linode 的計畫（畢竟先 dockerize 跑起來再說）
2. rumors-deploy 開一個 branch 把 docker-compose 改 helm chart
3. 試著把 staging 改成用 k8s deploy
4. 再想辦法如何弄 production
或許開個 branch 把落落長 docker-compose 的一兩個變成 helm chart yml 是個好開始～？這樣其他的 service 也能依樣畫葫蘆～～
deep fake 什麼的等出現了再來煩惱吧
don’t get distracted from replying rumors
Replied to a thread: 2021-05-24 13:51:46
關於 LINE bot 轉移到 Linode 的進度
請大家看一下這個 rumors-deploy 的 PR
之後預計會把 redis 與 LINE bot 搬到 Linode 上來作業
包含 LINE bot chat log 在內的 docker image stdout 會直接透過 docker engine 的 AWS log driver 送到 AWS cloudwatch（懶得用 vector 啦）
預計大小方面，五月開始到現在的對話紀錄大小為 1GB。Cloudwatch 有 5GB 的 free tier，大概五個月之後會開始付錢 (?)
請大家看一下這個 rumors-deploy 的 PR
之後預計會把 redis 與 LINE bot 搬到 Linode 上來作業
包含 LINE bot chat log 在內的 docker image stdout 會直接透過 docker engine 的 AWS log driver 送到 AWS cloudwatch（懶得用 vector 啦）
預計大小方面，五月開始到現在的對話紀錄大小為 1GB。Cloudwatch 有 5GB 的 free tier，大概五個月之後會開始付錢 (?)
- 👀1
Log into Facebook to start sharing and connecting with your friends, family, and people you know.
Facebook
前幾天有鎖朋友請大家注意，總之大概5/10組織，5/14開始攻勢，預計這一波會到5/28。攻勢密集是什麼意思呢？就是你平常備份的帳號、網站等等，都集中在同一個時候發動類似的內容。平常分享動物的專頁開始製造疫情假消息，假的蘋果與自由的網站開始運作，你大概就知道對方花了半年時間養的哪幾條線開始動了。至於沒動的線，你大概就知道他們彼此不是一夥的，或者預算有限。
就像我前幾天說的，在地協力者原本就很多，但是你很難分辨一般的受害者轉發、親中者，以及在地協力者之間的差別，所以這一兩年以來我們一直努力的就是用金流、人事、行為模式等等方式去比對。在這個容易對立的時刻，還是請大家不要隨便扣上他人是在地協力者的帽子。戒之慎之。
按照這個趨勢（假新聞網站、農場、在地錄音等），畢竟都是之前選舉時常見的集中攻擊，再來假公文、假情報就會隨之而來了。不能闢謠的謠言會越來越多，這時也必須仰賴大家的智慧了。先去忙了。
就像我前幾天說的，在地協力者原本就很多，但是你很難分辨一般的受害者轉發、親中者，以及在地協力者之間的差別，所以這一兩年以來我們一直努力的就是用金流、人事、行為模式等等方式去比對。在這個容易對立的時刻，還是請大家不要隨便扣上他人是在地協力者的帽子。戒之慎之。
按照這個趨勢（假新聞網站、農場、在地錄音等），畢竟都是之前選舉時常見的集中攻擊，再來假公文、假情報就會隨之而來了。不能闢謠的謠言會越來越多，這時也必須仰賴大家的智慧了。先去忙了。
把文字貼上來
PUMA老師揭露目前不實訊息最新發展
- 😮2
？？？
`不實訊息最新發展` ？
前幾天有鎖朋友請大家注意，總之大概5/10組織，5/14開始攻勢，預計這一波會到5/28。攻勢密集是什麼意思呢？就是你平常備份的帳號、網站等等，都集中在同一個時候發動類似的內容。平常分享動物的專頁開始製造疫情假消息，假的蘋果與自由的網站開始運作，你大概就知道對方花了半年時間養的哪幾條線開始動了。至於沒動的線，你大概就知道他們彼此不是一夥的，或者預算有限。
就像我前幾天說的，在地協力者原本就很多，但是你很難分辨一般的受害者轉發、親中者，以及在地協力者之間的差別，所以這一兩年以來我們一直努力的就是用金流、人事、行為模式等等方式去比對。在這個容易對立的時刻，還是請大家不要隨便扣上他人是在地協力者的帽子。戒之慎之。
按照這個趨勢（假新聞網站、農場、在地錄音等），畢竟都是之前選舉時常見的集中攻擊，再來假公文、假情報就會隨之而來了。不能闢謠的謠言會越來越多，這時也必須仰賴大家的智慧了。先去忙了。
就像我前幾天說的，在地協力者原本就很多，但是你很難分辨一般的受害者轉發、親中者，以及在地協力者之間的差別，所以這一兩年以來我們一直努力的就是用金流、人事、行為模式等等方式去比對。在這個容易對立的時刻，還是請大家不要隨便扣上他人是在地協力者的帽子。戒之慎之。
按照這個趨勢（假新聞網站、農場、在地錄音等），畢竟都是之前選舉時常見的集中攻擊，再來假公文、假情報就會隨之而來了。不能闢謠的謠言會越來越多，這時也必須仰賴大家的智慧了。先去忙了。
把文字貼上來
2021-05-25
@maofanhsu has joined the channel
@chelswdev has joined the channel
@kl310195 has joined the channel
2021-05-26
⚠️ *Action required - upgrade app for g0v* ⚠️
@acerxp511 @lucien line bot docker 化
staging 上現在跑的就是 build 出來的版本
staging 上現在跑的就是 build 出來的版本
@mrorz linode k8s?
cool 原來 linode 也有 k8s
雖然現在我們還是 docker-compose 打天下
我已經用很久了，@mrorz 搭配 github package 超好用
@mrorz 需要我幫你換 k8s 嗎
Linode k8s 的好處是 master 是不用錢的，只要付下面 node 費用😂😂
建議開三台，但是我省錢只開一台😂😂
聽起來很讚
所以是 rumors-deploy https://github.com/cofacts/rumors-deploy 的 docker-compose.yml 會變成 k8s 的 config 這樣嗎～～
所以是 rumors-deploy https://github.com/cofacts/rumors-deploy 的 docker-compose.yml 會變成 k8s 的 config 這樣嗎～～
看要不要直上 helm chart
他就會變 helm chart yaml
我只要管付錢跟開多大，剩下的 k8s 直接算這樣嗎
cooool
Db 的話 他會幫你開 linode storage
Mysql 會開 pod ，然後 pvc 會去掛 linode storage
我們是 elasticsearch
他有點肥
他有點肥
@mrorz 實際上就是你原本的 linode 花費
喔如果一台的話這樣嗎
只是想成不用自己進去管機器
變成管 k8s
那 cronjob 那些也是塞進 k8s ㄇ
我們現在還是有對 file system 做些髒髒事情
我們現在還是有對 file system 做些髒髒事情
例如說 bind mount 之類
這個要看你做了哪些😂
Cronjob的話 k8s 就 job scheduler
是說如果只有一台機器的話，那跟 docker-compose 差在哪裡 XD
雖然我是滿想拆小機器的
變成三台 20USD 的機器
變成三台 20USD 的機器
你可以開3台 20usd
但拆半天也就是省 20USD
因為我現在是一台 80USD
給 linode k8s 啊
對我來說 用 k8s 要加新服務或搬去其他 k8s 都很快啦
不過我個人目前是用 ko 做 deploy (所以一定要寫 golang lol)
感覺比較合適的做法會是
1. 先繼續把 rumors-line-bot 開在 linode 的計畫（畢竟先 dockerize 跑起來再說）
2. rumors-deploy 開一個 branch 把 docker-compose 改 helm chart
3. 試著把 staging 改成用 k8s deploy
4. 再想辦法如何弄 production
1. 先繼續把 rumors-line-bot 開在 linode 的計畫（畢竟先 dockerize 跑起來再說）
2. rumors-deploy 開一個 branch 把 docker-compose 改 helm chart
3. 試著把 staging 改成用 k8s deploy
4. 再想辦法如何弄 production
或許開個 branch 把落落長 docker-compose 的一兩個變成 helm chart yml 是個好開始～？這樣其他的 service 也能依樣畫葫蘆～～
- 3
看來「我想補充」設計果然還是要調整
嘛，說不定作者是真的想這樣回，所以他對功能的理解可能沒錯 XD
看來「我想補充」設計果然還是要調整
嘛，說不定作者是真的想這樣回，所以他對功能的理解可能沒錯 XD
最近幾次回報訊息的時候，特別是長的訊息，在「請選擇比較接近的版本」的時候，會有種「啊我沒辦法充分比較兩者異同，想看 diff」的感覺⋯
所以最後都是抖抖的按下第一個「選擇這篇」
咦，相似度與 highlight 沒有幫助嗎
（highlight 在 server 這裡好像吐得有點片段，這可以再修）
```【印度的變種病毒】
印度工作的中國人，由印度經尼泊爾返回國內，抵達重慶市時，初部檢測都是陰性.
但醫生仍有懷疑，於是幫他們進行CT電腦掃描，發現他們的肺部有病變，並確診是印度的三重變種病毒,証明這種印度超級變種病毒，可以避過現在的檢測,而大部份的關口都只是靠一般普通測試
而香港及大部份的國家，不會強制入境人士進行CT掃描,亦沒有足夠時間及資源去做。
如果估計正確、相信第五波超級病毒大爆發，即將出現在香港及全球！
自新加坡傳來的訊息：*病毒回來了*。
這次有更強的精力、戰術和偽裝。感染者不咳嗽，沒有發燒。
*這次症狀是關節疼痛、無力、食慾不振。*
死亡率更高，達到危急的時間更短。有時沒有症狀...要小心...！
該病毒株不在我們的鼻咽區域內匿居！所以不再有嗅覺或味覺喪失等前驅症狀，它直接攻擊肺部，縮短了發病時間。
許多不發燒的患者，但是X光檢查顯示中度胸腔肺炎！鼻黏膜篩試常顯示COVID19陰性！越來越多的假性咽喉鼻腔檢查結果。
這意味著病毒直接傳播到肺部，由病毒性肺炎而引起急性呼吸窘迫(缺氧)！
這解釋了為什麼它變得急性和致命得多！發燒時可能已變成重症。
請注意:*避免到擁擠的地方*，*保持社交距離*，*戴上口罩*，*經常用洗手乳或肥皂洗手*。這次疫情比上一波更致命，必須小心謹慎，*不可輕忽*！```
印度工作的中國人，由印度經尼泊爾返回國內，抵達重慶市時，初部檢測都是陰性.
但醫生仍有懷疑，於是幫他們進行CT電腦掃描，發現他們的肺部有病變，並確診是印度的三重變種病毒,証明這種印度超級變種病毒，可以避過現在的檢測,而大部份的關口都只是靠一般普通測試
而香港及大部份的國家，不會強制入境人士進行CT掃描,亦沒有足夠時間及資源去做。
如果估計正確、相信第五波超級病毒大爆發，即將出現在香港及全球！
自新加坡傳來的訊息：*病毒回來了*。
這次有更強的精力、戰術和偽裝。感染者不咳嗽，沒有發燒。
*這次症狀是關節疼痛、無力、食慾不振。*
死亡率更高，達到危急的時間更短。有時沒有症狀...要小心...！
該病毒株不在我們的鼻咽區域內匿居！所以不再有嗅覺或味覺喪失等前驅症狀，它直接攻擊肺部，縮短了發病時間。
許多不發燒的患者，但是X光檢查顯示中度胸腔肺炎！鼻黏膜篩試常顯示COVID19陰性！越來越多的假性咽喉鼻腔檢查結果。
這意味著病毒直接傳播到肺部，由病毒性肺炎而引起急性呼吸窘迫(缺氧)！
這解釋了為什麼它變得急性和致命得多！發燒時可能已變成重症。
請注意:*避免到擁擠的地方*，*保持社交距離*，*戴上口罩*，*經常用洗手乳或肥皂洗手*。這次疫情比上一波更致命，必須小心謹慎，*不可輕忽*！```
喔喔喔喔這超眼熟 XD
這張圖可以借我放 github issue 嗎 XD
i see
其實第一篇應該幾乎全都 highlight 起來才對
如果我們修正那個 highlight，或許就有用
其實第一篇應該幾乎全都 highlight 起來才對
如果我們修正那個 highlight，或許就有用
- 🙌1
好像是欸
可能有喔 \/
\o/
不過我還是想看到左邊右邊兩欄的 diff（請忽略）
其實 google search 也只有 highlight 而沒 diff
但我們也能選出一個我們想點進去的
chatbot 這個情境跟 google 的情境有什麼不同，會讓你想到 diff 呢
但我們也能選出一個我們想點進去的
chatbot 這個情境跟 google 的情境有什麼不同，會讓你想到 diff 呢
highlight 相關
https://github.com/cofacts/rumors-api/issues/229
https://github.com/cofacts/rumors-api/issues/229
- 👍1
Replied to a thread: 2021-05-26 18:05:18
這張圖可以借我放 github issue 嗎 XD
沒問題 XD 要我 comment 也可以
Replied to a thread: 2021-05-26 18:06:48
其實 google search 也只有 highlight 而沒 diff
但我們也能選出一個我們想點進去的
chatbot 這個情境跟 google 的情境有什麼不同，會讓你想到 diff 呢
但我們也能選出一個我們想點進去的
chatbot 這個情境跟 google 的情境有什麼不同，會讓你想到 diff 呢
有道理
沒有理論基礎的猜測：如果看到開頭不一樣，就會覺得好像不一樣
<https://g0v-slack-archive.g0v.ronny.tw/index/channel/C2PPMRQGP#ts-1622023244.0592|https://g0v-slack-archive.g0v.ronny.tw/index/channel/C2PPMRQGP#ts-1622023244.0592> Text: > 【印度的變種病毒】 > 印度工作的中國人，由印度經尼泊爾返回國內，抵達重慶市時，初部檢測都是陰性. > 但醫生仍有懷疑，於是幫他們進行CT電腦掃描，發現他們的肺部有病變，並確診是印度的三重變種病毒,証明這種印度超級變種病毒，可以避過現在的檢測,而大部份的關口都只是靠一般普通測試 > 而香港及大部份的國家，不會強制入境人士進行CT掃描,亦沒有足夠時間及資源去做。 > 如果估計正確、相信第五波超級病毒大爆發，即將出現在香港及全球！ > 自新加坡傳來的訊息：_病毒回來了_。 > 這次有更強的精力、戰術和偽裝。感染者不咳嗽，沒有發燒。 > _這次症狀是關節疼痛、無力、食慾不振。_ > 死亡率更高，達到危急的時間更短。有時沒有症狀...要小心...！ > 該病毒株不在我們的鼻咽區域內匿居！所以不再有嗅覺或味覺喪失等前驅症狀，它直接攻擊肺部，縮短了發病時間。 > 許多不發燒的患者，但是X光檢查顯示中度胸腔肺炎！鼻黏膜篩試常顯示COVID19陰性！越來越多的假性咽喉鼻腔檢查結果。 > 這意味著病毒直接傳播到肺部，由病毒性肺炎而引起急性呼吸窘迫(缺氧)！ > 這解釋了為什麼它變得急性和致命得多！發燒時可能已變成重症。 > 請注意:_避免到擁擠的地方_，_保持社交距離_，_戴上口罩_，_經常用洗手乳或肥皂洗手_。這次疫情比上一波更致命，必須小心謹慎，_不可輕忽_！ Highlight: <https://user-images.githubusercontent.com/108608/119642692-9070c180-be4d-11eb-8fb1-7a02ef03a239.png|image> Improvements: 1. All matching text should be highlighted 2. Should show high density of highlights
- 🙌1
然後就覺得：如果有個更全面的比較視覺化，應該可以更確定是不是這一則！然後 diff 只是因為那是一種習慣的介面所以就直覺想到它
了解了解
- 🙌1
highlight improvement 是比較有明確 todo item 的部分
不過謠言的段落拼貼，如果可以視覺化好像也滿清楚的，可以知道哪裡「是加的」
是個有趣的想法～～感謝感謝
不過謠言的段落拼貼，如果可以視覺化好像也滿清楚的，可以知道哪裡「是加的」
是個有趣的想法～～感謝感謝
- 🙌1
雖然我想到的視覺化，通常要有足夠的螢幕空間才有辦法呈現呢
小空間大挑戰 +1 q_q
@yi94025 has joined the channel
2021-05-27
@side112358 has joined the channel
@6g2wqhhvpn has joined the channel
2021-05-28
@ypc9314 has joined the channel
2021-05-29
@terriecychang has joined the channel
2021-05-30
@terry.f.wang has joined the channel
@minexo79 has joined the channel
@hyy6983 has joined the channel
⚠️ *Action required - upgrade app for g0v* ⚠️
In this PR, we make most keys volatile and to `lastScannedAt` key is persistent. After deploy, we can change Redis `maxmemory-policy` from `allkeys-lru` to `volatile-lru` to ensure that `maxmemory-policy` is never evicted. This PR also handles the case when postback actions are called after the search session is removed from Redis. This should fix <https://rollbar.com/mrorz/rumors-line-bot/items/220/|https://rollbar.com/mrorz/rumors-line-bot/items/220/> .
2021-05-31
@hsiangtai.chien has joined the channel
@kelvinho84 has joined the channel
@aurora0402 has joined the channel
@lipuyeh has joined the channel
@yenchih14 has joined the channel
@ian72722000 has joined the channel