Before

The CSS properties are what also known as . They are used to add something before or after the content of an element. There are a lot of great uses for these pseudo elements, and we are here to explore some of them. :before and :after pseudo elements The Syntax If we have an element like this one: welcome to our website < > h2 </ > h2 We can add a pseudo element before it using CSS, like this: { : “Hello“; : blue;
} h2 :before content color The result will be: This is a quite simple principle. You are adding content before or after a certain element. It can be extremely helpful when , and in many other cases. adding icons, clearing floats The property of the pseudo element can be left with empty quotes like this .  This way you can treat the pseudo element like a . If the content property is removed altogether, the pseudo element content empty `content: “”` box with no content will not work. Adding Icons Probably the most popular usage of the before and after pseudo elements is using them to . We can add a simple .png icon. add icons Let’s look at the markup. HTML Weather report < > h2 </ > h2 CSS { : “ ”; : block; : ; : ; : (‘images/icon.png’’); : ; h2 :before content display height 15px width 25px url margin-right 5px The result: Now we have successfully added an icon before the text. Easy, right? Clearing Floats After elements are , another one needs to be added in order to clear that float. You can and simply using a pseudo one. floated avoid adding a new element Let's say we have this situation: We can use the next code to achieve clearing of the floats. HTML Lorem ipsum dolor amet truffaut kombucha roof party iPhone forage farm-to-table. < = > div class "box-container" < = > div class ”box” </ > div < = > div class ”box” </ > div </ > div < > p </ > p CSS , { : ; : block;
} { : both;
} { : ; : ; : ; : ; : left;
} .box-container :before .box-container :after content "" display .box-container :after clear .box height 100px width 100px background-color #C98EED margin 5px float Now, let's take a look at the result. By using this method we are and the paragraph is moved below the two elements. clearing the float Using a Background Image We can also to a pseudo element. This is commonly used when styling a header. add a background image HTML Hello World < > h2 </ > h2 CSS { : “”; : ; : ; : (‘underline.png’) center center no-repeat; : block;
} h2 :after content width 100% height 30px background url display The result achieved: Browser Support As with everything else in CSS, browser support needs to be checked. By consulting the , we see that these pseudo classes have a high browser support ( ) and we won’t have a headache when using them. CanIUse over 98% Summary Here, we have explained of the CSS pseudo elements. The examples illustrate just some of the many possible usages. Don't worry if it's not completely clear in the beginning. A little practice goes a long way. basic principles Hopefully, this article will help with any future projects. Thank you for reading! Previously published at https://kolosek.com/css-before-and-after/

How to use CSS pseudo-classes :before and :after

In RSpec, there are two different ways to write DRY tests, by using or . Their purpose is to create variables that are common across tests. In this post, we will explore differences between before and let and explain . before let why let is preferred by the Ruby community let Let creates lazily-evaluated local variables. This means that let() is not evaluated until the method that it defines is run for the first time. It DRYs up the spec and makes it more readable. $count = describe let( ) { $count += }
  it expect(count).to eq( )
    expect(count).to eq( ) it expect(count).to eq( ) 0 "let" do :count 1 "stores the value" do 1 1 end "is not cached across examples" do 2 end end Let should not be used for local variables, which have to be saved to the database, as they will not be saved to the database unless they have already been referenced. In this case, you should use or . let! before blocks The the tests we are creating are done using . RSpec and Capybara Also, have a let block inside of a before the block, this is what let! is made for! never let! Unlike let, you can . It means that, even if you didn't invoke the helper method inside the example, it will be invoked before your example runs. use let! to force the method's invocation before each example $count = describe invocation_order = []
  let!( ) invocation_order << $count += it invocation_order << expect(invocation_order).to eq([ , ])
    expect(count).to eq( ) 0 "let!" do :count do :let! 1 end "calls the helper method in a before hook" do :example :let! :example 1 end end As with let blocks, if multiple let! blocks are defined with the same name, the most recent one will be executed. The core difference is that will be executed multiple times if used like this, whereas the will only execute the last time. let! blocks let block before(:each) Before(:each) block will run before each example, even if the example doesn't use any of the instance variables defined in the block. This can noticeably slow down the setup of the instance variables. @tests = [] describe User before( ) @user = User.new describe it @user.should have( ).tests it @user.tests << Object.new it @user.should have( ).tests class User def tests || end end do :each do end "initialized in before(:each)" do "has 0 tests" do 0 end "can accept new tests" do end "does not share state across examples" do 0 end end end Depending on your personal preference you could use before blocks when: In nearly every situation, it is better to use let over before blocks. There is a reasonable amount of variables. There are variables that don't need to be referenced directly but are required. There are many commands to be executed because its syntax is more clear when many commands are involved. Creating mocks/stubs. Give it a try and create your own model or ! controller tests before(:all) This block is executed only once, before all of the examples in a group. There are certain situations this can cut down on execution and effort. @tests = [] describe User before( ) @user = User.new describe it @user.should have( ).tests it @user.tests << Object.new it @user.should have( ).tests class User def tests || end end do :all do end "initialized in before(:all)" do "has 0 tests" do 0 end "can get accept new tests" do end "shares state across examples" do 1 end end end It runs outside of transactions, so the data created here will bleed into other specs. Using before(:all) in RSpec will cause you lots of trouble unless you know what you are doing! Conclusion It all depends on what and how you need to make the work or consider using for creating data. Let blocks bring more to the table than before blocks. RSpec tests FactoryGirl Besides being slower, one of the major problems with before blocks is that spelling errors can lead to bugs and false positives, allowing certain types of tests to pass when they shouldn't. before( ) @user = User.find( )
    @user.logout it expect(@usr).to be_nil :each do username: "kolosek" end "should log the user out" do end Since @usr wasn't previously defined, the test will pass, @usr is nil by default. The same test using let would raise NameError because @usr isn't defined. Hope this will help you to better understand the differences between let and before blocks. Thank you for reading! Previously published at https://kolosek.com/rspec-let-vs-before/

RSpec Testing: Let Vs. Before

PALO ALTO, Calif., USA, December 30th, 2024/CyberNewsWire/--SquareX, an industry-first Browser Detection and Response (BDR) solution, leads the way in browser security. About a week ago, SquareX reported large-scale attacks targeting Chrome Extension developers aimed at taking over the Chrome Extension from the Chrome Store. On December 25th, 2024, a malicious version of Cyberhaven’s browser extension was published on the Chrome Store that allowed the attacker to hijack authenticated sessions and exfiltrate confidential information. The malicious extension was available for download for more than 30 hours before being removed by Cyberhaven. The data loss prevention company declined to comment on the extent of the impact when approached by the press, but the extension had over 400,000 users on the Chrome Store at the time of the attack. Unfortunately, the attack took place as SquareX's researchers had identified a similar attack with a video demonstrating the entire attack pathway just a week before the Cyberhaven breach. The attack begins with a phishing email impersonating Chrome Store containing a supposed violation of the platform’s “Developer Agreement”, urging the receiver to accept the policies to prevent their extension from being removed from Chrome Store. Upon clicking on the policy button, the user gets prompted to connect their Google account to a “Privacy Policy Extension”, which grants the attacker access to edit, update and publish extensions on the developer’s account. Fig 1. Phishing email targeting extension developers Fig 2. Fake Privacy Policy Extension requesting access to “edit, update or publish” the developer’s extension Extensions have become an increasingly popular way for attackers to gain initial access. This is because most organizations have limited purview on what browser extensions their employees are using. Even the most rigorous security teams typically do not monitor subsequent updates once an extension is whitelisted. SquareX has conducted extensive research and demonstrated at DEFCON 32, how MV3-compliant extensions can be used to steal video stream feeds, add a silent GitHub collaborator, and steal session cookies, among others. Attackers can create a seemingly harmless extension and later convert it into a malicious one post-installation or, as demonstrated in the attack above, deceive the developers behind a trusted extension to gain access to one that already has hundreds of thousands of users. In Cyberhaven’s case, attackers were able to steal company credentials across multiple websites and web apps through the malicious version of the extension. Given that developer emails are publicly listed on Chrome Store, it is easy for attackers to target thousands of extension developers at once. These emails are typically used for bug reporting. Thus, even support emails listed for extensions from larger companies are usually routed to developers who may not have the level of security awareness required to find suspicion in such an attack. As per SquareX’s attack disclosure and the Cyberhaven breach that occurred within the span of less than two weeks, the company has strong reason to believe that many other browser extension providers are being attacked in the same way. SquareX urges companies and individuals alike to conduct a careful inspection before installing or updating any browser extensions. Fig 3. Contact details of extension developers are publicly available on Chrome Store SquareX team understands that it can be non-trivial to evaluate and monitor every single browser extension in the workforce amidst all the competing security priorities, especially when it comes to zero-day attacks. As demonstrated in the video, the fake privacy policy app involved in Cyberhaven’s breach was not even detected by any popular threat feeds. SquareX’s Browser Detection and Response (BDR) solution takes this complexity off security teams by: Blocking OAuth interactions to unauthorized websites to prevent employees from accidentally giving attackers unauthorized access to your Chrome Store account
Blocking and/or flagging any suspicious extension updates containing new, risky permissions
Blocking and/or flagging any suspicious extensions with a surge of negative reviews
Blocking and/or flagging installations of sideloaded extensions
Streamline all requests for extension installations outside the authorized list for quick approval based on company policy
Full visibility on all extensions installed and used by employees across the organization SquareX’s founder Vivek Ramachandran warns: “Identity attacks targeting browser extensions similar to this OAuth attack will only become more prevalent as employees rely on more browser-based tools to be productive at work. Similar variants of these attacks have been used in the past to steal cloud data from apps like Google Drive and One Drive and we will only see attackers get more creative in exploiting browser extensions. Companies need to remain vigilant and minimize their supply chain risk without hampering employee productivity by equipping them with the right browser native tools.” About SquareX: SquareX helps organizations detect, mitigate, and threat-hunt client-side web attacks happening against their users in real-time. SquareX's industry-first Browser Detection and Response (BDR) solution, takes an attack-focused approach to browser security, ensuring enterprise users are protected against advanced threats like malicious QR Codes, Browser-in-the-Browser phishing, macro-based malware, and other web attacks encompassing malicious files, websites, scripts, and compromised networks. With SquareX, enterprises can provide contractors and remote workers with secure access to internal applications, and enterprise SaaS, and convert the browsers on BYOD / unmanaged devices into trusted browsing sessions. Contact Head of PR Junice Liew SquareX junice@sqrx.com This story was distributed as a release by Cybernewswire under HackerNoon’s Business Blogging Program. Learn more about the program here PALO ALTO, Calif., USA, December 30th, 2024/CyberNewsWire/--SquareX, an industry-first Browser Detection and Response (BDR) solution, leads the way in browser security. About a week ago, SquareX reported large-scale attacks targeting Chrome Extension developers aimed at taking over the Chrome Extension from the Chrome Store. reported reported On December 25th, 2024, a malicious version of Cyberhaven’s browser extension was published on the Chrome Store that allowed the attacker to hijack authenticated sessions and exfiltrate confidential information. The malicious extension was available for download for more than 30 hours before being removed by Cyberhaven. The data loss prevention company declined to comment on the extent of the impact when approached by the press, but the extension had over 400,000 users on the Chrome Store at the time of the attack. malicious extension malicious extension Chrome Store Chrome Store Unfortunately, the attack took place as SquareX's researchers had identified a similar attack with a video demonstrating the entire attack pathway just a week before the Cyberhaven breach. identified identified video video The attack begins with a phishing email impersonating Chrome Store containing a supposed violation of the platform’s “Developer Agreement”, urging the receiver to accept the policies to prevent their extension from being removed from Chrome Store. Upon clicking on the policy button, the user gets prompted to connect their Google account to a “Privacy Policy Extension”, which grants the attacker access to edit, update and publish extensions on the developer’s account. Fig 1. Phishing email targeting extension developers Fig 2. Fake Privacy Policy Extension requesting access to “edit, update or publish” the developer’s extension Extensions have become an increasingly popular way for attackers to gain initial access. This is because most organizations have limited purview on what browser extensions their employees are using. Even the most rigorous security teams typically do not monitor subsequent updates once an extension is whitelisted. SquareX has conducted extensive research and demonstrated at DEFCON 32, how MV3-compliant extensions can be used to steal video stream feeds, add a silent GitHub collaborator, and steal session cookies, among others. DEFCON 32, DEFCON 32, Attackers can create a seemingly harmless extension and later convert it into a malicious one post-installation or, as demonstrated in the attack above, deceive the developers behind a trusted extension to gain access to one that already has hundreds of thousands of users. In Cyberhaven’s case, attackers were able to steal company credentials across multiple websites and web apps through the malicious version of the extension. Given that developer emails are publicly listed on Chrome Store, it is easy for attackers to target thousands of extension developers at once. These emails are typically used for bug reporting. Thus, even support emails listed for extensions from larger companies are usually routed to developers who may not have the level of security awareness required to find suspicion in such an attack. As per SquareX’s attack disclosure and the Cyberhaven breach that occurred within the span of less than two weeks, the company has strong reason to believe that many other browser extension providers are being attacked in the same way. SquareX urges companies and individuals alike to conduct a careful inspection before installing or updating any browser extensions. Fig 3. Contact details of extension developers are publicly available on Chrome Store SquareX team understands that it can be non-trivial to evaluate and monitor every single browser extension in the workforce amidst all the competing security priorities, especially when it comes to zero-day attacks. As demonstrated in the video , the fake privacy policy app involved in Cyberhaven’s breach was not even detected by any popular threat feeds. video video SquareX’s Browser Detection and Response (BDR) solution takes this complexity off security teams by: SquareX’s Browser Detection and Response (BDR) solution SquareX’s Browser Detection and Response (BDR) solution Blocking OAuth interactions to unauthorized websites to prevent employees from accidentally giving attackers unauthorized access to your Chrome Store account Blocking and/or flagging any suspicious extension updates containing new, risky permissions Blocking and/or flagging any suspicious extensions with a surge of negative reviews Blocking and/or flagging installations of sideloaded extensions Streamline all requests for extension installations outside the authorized list for quick approval based on company policy Full visibility on all extensions installed and used by employees across the organization Blocking OAuth interactions to unauthorized websites to prevent employees from accidentally giving attackers unauthorized access to your Chrome Store account Blocking and/or flagging any suspicious extension updates containing new, risky permissions Blocking and/or flagging any suspicious extensions with a surge of negative reviews Blocking and/or flagging installations of sideloaded extensions Streamline all requests for extension installations outside the authorized list for quick approval based on company policy Full visibility on all extensions installed and used by employees across the organization SquareX’s founder Vivek Ramachandran warns: “Identity attacks targeting browser extensions similar to this OAuth attack will only become more prevalent as employees rely on more browser-based tools to be productive at work. Similar variants of these attacks have been used in the past to steal cloud data from apps like Google Drive and One Drive and we will only see attackers get more creative in exploiting browser extensions. Companies need to remain vigilant and minimize their supply chain risk without hampering employee productivity by equipping them with the right browser native tools.” SquareX’s founder Vivek Ramachandran warns: “Identity attacks targeting browser extensions similar to this OAuth attack will only become more prevalent as employees rely on more browser-based tools to be productive at work. Similar variants of these attacks have been used in the past to steal cloud data from apps like Google Drive and One Drive and we will only see attackers get more creative in exploiting browser extensions. Companies need to remain vigilant and minimize their supply chain risk without hampering employee productivity by equipping them with the right browser native tools.” Vivek Ramachandran Vivek Ramachandran About SquareX: SquareX helps organizations detect, mitigate, and threat-hunt client-side web attacks happening against their users in real-time. SquareX SquareX SquareX's industry-first Browser Detection and Response (BDR) solution, takes an attack-focused approach to browser security, ensuring enterprise users are protected against advanced threats like malicious QR Codes, Browser-in-the-Browser phishing, macro-based malware, and other web attacks encompassing malicious files, websites, scripts, and compromised networks. With SquareX, enterprises can provide contractors and remote workers with secure access to internal applications, and enterprise SaaS, and convert the browsers on BYOD / unmanaged devices into trusted browsing sessions. Contact Head of PR Junice Liew SquareX junice@sqrx.com This story was distributed as a release by Cybernewswire under HackerNoon’s Business Blogging Program. Learn more about the program here This story was distributed as a release by Cybernewswire under HackerNoon’s Business Blogging Program. Learn more about the program here here here

SquareX Researchers Expose OAuth Attack On Chrome Extensions Days Before Major Breach

Даследчыкі SquareX выкрылі атаку OAuth на пашырэнні Chrome за некалькі дзён да сур'ёзнага ўзлому

Investigadores de SquareX descubren un ataque OAuth en extensiones de Chrome días antes de una importante vulneración

Istraživači SquareX-a razotkrivaju OAuth napad na Chromeova proširenja danima prije većeg proboja

חוקרי SquareX חושפים את התקפת OAuth על הרחבות Chrome ימים לפני הפרה גדולה

SquareX の研究者が、大規模な侵害の数日前に Chrome 拡張機能への OAuth 攻撃を暴露

SquareX зерттеушілері Chrome кеңейтімдеріндегі OAuth шабуылын үлкен бұзылулардан бірнеше күн бұрын ашты

SquareX 연구원, 주요 침해 발생 며칠 전 크롬 확장 프로그램에 대한 OAuth 공격 폭로

SquareX изилдөөчүлөрү Chrome кеңейтүүлөрүнө OAuth чабуулун ири бузуулардан бир нече күн мурун ашкерелешти

ນັກຄົ້ນຄວ້າ SquareX ເປີດເຜີຍການໂຈມຕີ OAuth ກ່ຽວກັບ Chrome Extensions ມື້ກ່ອນການລະເມີດທີ່ສໍາຄັນ

Badacze SquareX ujawniają atak OAuth na rozszerzenia Chrome na kilka dni przed poważnym naruszeniem

SquareX-forskare avslöjar OAuth-attack på Chrome-tillägg dagar innan större intrång

Abaphandi be-SquareX Baveza uhlaselo lwe-OAuth kwiiNtsuku zoLwandiso lwe-Chrome ngaphambi kokwaphulwa okukhulu

Analog, a blockchain protocol focusing on cross-chain communication, launched its network today, enabling connections between different blockchain platforms. The initial release operates on a Proof-of-Authority mechanism, with plans to transition to Nominated Proof of Stake by the time of its token release. The network launch follows testing phases that processed 1.9 million cross-chain data transfers among 345,000 participating accounts. This testing period allowed developers to examine the protocol's capabilities and provide technical feedback before the main release. "Cross-chain communication remains a central challenge in blockchain development," says Dr. Elena Martinez, blockchain researcher at MIT's Digital Currency Institute. "Solutions that enable different chains to interact could reduce fragmentation in the blockchain space." The protocol's architecture includes components for processing data between chains, executing smart contracts across networks, and maintaining consistent information across systems. These features aim to address current limitations in blockchain interoperability, where applications often operate in isolation on single networks. Several blockchain projects have begun implementing the protocol. Frax Finance, which operates a decentralized stablecoin system, plans to use the network for transmitting price data across different blockchain platforms. This implementation could help maintain consistent asset pricing across multiple networks. XYO Network, which operates location-tracking systems on blockchain, intends to use the protocol to make geographical data accessible across different chains. This application demonstrates potential uses beyond financial transactions, extending to real-world data integration. The protocol also enables governance activities across chains. Token holders of various projects can participate in voting and decision-making regardless of which blockchain holds their assets, potentially increasing participation in project governance. Technical metrics from the testing period indicate sustained usage, with users listing 290,000 smart contracts across multiple chains. The protocol processed these contracts while maintaining data consistency, suggesting capability for sustained operation at scale. Security measures include identity verification systems, with 42,000 users completing verification processes before participating. These measures aim to maintain network integrity while enabling broad participation. Looking forward, the development team plans additional features, including the consensus mechanism transition and expanded cross-chain messaging capabilities. These updates aim to increase the network's capability for handling complex multi-chain operations. The launch represents progress in blockchain interoperability research, potentially contributing to increased communication between previously isolated networks. As blockchain systems continue operating on separate networks, protocols enabling cross-chain interaction may become increasingly relevant for developers building distributed applications. Developers interested in examining the protocol's capabilities can access technical documentation through official channels, with the network now processing live transactions and cross-chain communications. Don’t forget to like and share the story! Vested Interest Disclosure: This author is an independent contributor publishing via our business blogging program. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR Analog , a blockchain protocol focusing on cross-chain communication, launched its network today, enabling connections between different blockchain platforms. The initial release operates on a Proof-of-Authority mechanism, with plans to transition to Nominated Proof of Stake by the time of its token release. Analog The network launch follows testing phases that processed 1.9 million cross-chain data transfers among 345,000 participating accounts. This testing period allowed developers to examine the protocol's capabilities and provide technical feedback before the main release. "Cross-chain communication remains a central challenge in blockchain development," says Dr. Elena Martinez, blockchain researcher at MIT's Digital Currency Institute. "Solutions that enable different chains to interact could reduce fragmentation in the blockchain space." The protocol's architecture includes components for processing data between chains, executing smart contracts across networks, and maintaining consistent information across systems. These features aim to address current limitations in blockchain interoperability, where applications often operate in isolation on single networks. Several blockchain projects have begun implementing the protocol. Frax Finance , which operates a decentralized stablecoin system, plans to use the network for transmitting price data across different blockchain platforms. This implementation could help maintain consistent asset pricing across multiple networks. Frax Finance XYO Network, which operates location-tracking systems on blockchain, intends to use the protocol to make geographical data accessible across different chains. This application demonstrates potential uses beyond financial transactions, extending to real-world data integration. The protocol also enables governance activities across chains. Token holders of various projects can participate in voting and decision-making regardless of which blockchain holds their assets, potentially increasing participation in project governance. Technical metrics from the testing period indicate sustained usage, with users listing 290,000 smart contracts across multiple chains. The protocol processed these contracts while maintaining data consistency, suggesting capability for sustained operation at scale. Security measures include identity verification systems, with 42,000 users completing verification processes before participating. These measures aim to maintain network integrity while enabling broad participation. Looking forward, the development team plans additional features, including the consensus mechanism transition and expanded cross-chain messaging capabilities. These updates aim to increase the network's capability for handling complex multi-chain operations. The launch represents progress in blockchain interoperability research, potentially contributing to increased communication between previously isolated networks. As blockchain systems continue operating on separate networks, protocols enabling cross-chain interaction may become increasingly relevant for developers building distributed applications. Developers interested in examining the protocol's capabilities can access technical documentation through official channels, with the network now processing live transactions and cross-chain communications. Don’t forget to like and share the story! Vested Interest Disclosure: This author is an independent contributor publishing via our business blogging program. HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR Vested Interest Disclosure: This author is an independent contributor publishing via our business blogging program . HackerNoon has reviewed the report for quality, but the claims herein belong to the author. #DYOR Vested Interest Disclosure: business blogging program business blogging program

Analog's Cross-Chain System Processes 1.9M Data Points Before Launch

Analoog se kruiskettingstelsel verwerk 1,9 miljoen datapunte voor bekendstelling

يقوم نظام Cross-Chain من Analog بمعالجة 1.9 مليون نقطة بيانات قبل الإطلاق

Analog ukax Sistema de Cadena Cruzada ukax 1.9M Puntos de Datos Janïr Lanzamiento ukaruw puri

Analoqun Çarpaz Zəncirli Sistemi işə salınmazdan əvvəl 1,9 milyon məlumat nöqtəsini emal edir

Analogov cross-chain sistem obrađuje 1,9 miliona tačaka podataka prije lansiranja

El sistema de cadena cruzada de Analog procesa 1,9 millones de puntos de datos antes del lanzamiento

מערכת חוצת השרשרת של Analog מעבדת 1.9 מיליון נקודות נתונים לפני ההשקה

アナログのクロスチェーンシステムは発売前に190万のデータポイントを処理

Analogā starpķēžu sistēma pirms palaišanas apstrādā 1,9 miljonus datu punktu

Analógový cross-chain systém pred spustením spracuje 1,9 milióna dátových bodov

Sistemi ndër-zinxhiror i Analog përpunon 1,9 milion pikë të dhënash përpara nisjes

Analogs tvärkedjesystem behandlar 1,9 miljoner datapunkter före lansering

Authors:
(1) Hanoona Rasheed, Mohamed bin Zayed University of AI and equally contributing first authors;
(2) Muhammad Maaz, Mohamed bin Zayed University of AI and equally contributing first authors;
(3) Sahal Shaji, Mohamed bin Zayed University of AI;
(4) Abdelrahman Shaker, Mohamed bin Zayed University of AI;
(5) Salman Khan, Mohamed bin Zayed University of AI and Australian National University;
(6) Hisham Cholakkal, Mohamed bin Zayed University of AI;
(7) Rao M. Anwer, Mohamed bin Zayed University of AI and Aalto University;
(8) Eric Xing, Mohamed bin Zayed University of AI and Carnegie Mellon University;
(9) Ming-Hsuan Yang, University of California - Merced and Google Research;
(10) Fahad S. Khan, Mohamed bin Zayed University of AI and Linköping University. Editor's Note: This is Part 4 of 10 of a study detailing the development of an AI model that is designed to describe images to users. Read the rest below. Table of Links Abstract and 1 Introduction
2. Related Work
3. Method
4. Data Annotation Pipeline
5. Experiments
6. Conclusion and References Supplementary Material (Part 1) A. Additional Implementation Details
B. Additional Downstream Tasks
C. Additional Qualitative Results Supplementary Material (Part 2) D. Dataset Visualization
E. Limitations and Future Work
F. Ethics and Societal Impact 4. Data Annotation Pipeline We introduce our automated annotation pipeline used to create the Grounding-anything Dataset (GranD). GranD is a comprehensive, multi-purpose image-text dataset offering a range of contextual information, from fine-grained to high-level details. It aims to overcome challenges in image understanding and dense pixel-level grounding, thereby expanding the capabilities of visual instruction tuning in LMMs. The pipeline contains four distinct levels (see Fig. 4). i) Level-1 focuses on object localization and provides semantic labels, segmentation masks, attributes, and depth information. ii) Level-2 defines relationships between detected objects. iii) Level-3 organizes information from the first two levels into a hierarchical scene graph, used to generate dense captions using LLM with in-context examples. iv) Level-4 offers enriched contextual information for a deeper understanding of the scene, going beyond what’s observed (e.g., historical information of a landmark). Please refer to Appendix A.4 for pipeline implementation details. 4.1. Object Localization and Attributes (Level-1) In level-1, the focus is on detailed object identification within images. First, object-bounding boxes are identified using multiple SoTA object detection models. Classagnostic NMS is applied to each model to filter out false positives. After this step, bounding boxes from different models are compared using IoU, with a bounding box retained as an object only if detected by at least two other detection models. We also generate attributes for each filtered object using region-based vision-language models and incorporate depth information to contextualize each object’s relative position within the scene. 4.2. Relationships and Landmarks (Level-2) In level-2, multiple short textual descriptions of the overall scene are generated. Phrases extracted from these descriptions are grounded to specific objects in level-1 to form relationships. These relationships articulate connections between multiple objects or define an object’s role within the scene. Further, each scene is assigned a landmark category that includes a primary and a more specific sub-category (see Tab. 7 in Appendix 7). 4.3. Scene Graph and Dense Captioning (Level-3) In level-3, object attributes and labels from level-1 are combined with the relationships and phrases obtained from level-2 to form a hierarchical scene graph. This structured data serves as a query for LLM to generate dense image captions. To provide additional context, depth values and bounding box coordinates are used to assign each object to specific spatial layers within the scene, such as immediate foreground, foreground, midground, or background. Additionally, short scene-level captions are incorporated into the scene graph to enhance LLMs’ contextual understanding. Dense Captioning Verification: To enhance the fidelity of the LLM-generated dense captions, we implement an automatic verification pipeline using chain-of-thoughts prompting. This pipeline produces a checklist of objects derived from the generated dense caption expected to be present in the image. The associated caption is flagged as inaccurate if any object specified in the checklist is absent from the scene graph. Such captions are then regenerated, incorporating feedback from the initial assessment. 4.4. Extra Contextual Insights (Level-4) Level-4 builds on the scene graph from level-3 to obtain a more detailed visual understanding. we query LLM to extract extended contextual insights beyond basic object identification and relationships, including details about the landmarks, historical context, guidelines for interacting with the scene, and even predictive elements about future events. To facilitate this, we prompt LLM with in-context examples. Utilizing our automated annotation pipeline, we annotate a corpus of 11M SAM images [18], which are inherently diverse, high-resolution, and privacy-compliant. The resulting dataset comprises 810M regions, each associated with a segmentation mask, and includes 7.5M unique concepts. Further, the dataset features 84M referring expressions, 22M grounded short captions, and 11M densely grounded captions. To our knowledge, this is the first dataset of this scale generated entirely through an automated annotation pipeline (see Tab. 2 for details and Fig. 15 for dataset sample visualizations). We extend open-source datasets—namely Flickr30K [37], RefCOCOg [16], and PSG [49] by generating compatible GCG annotations. For RefCOCOg, we use the dataset’s referring expressions and their connected masks. These expressions offer concise descriptions of distinct objects in the image. With the aid of GPT-4, we seamlessly blend these referring expressions with contextual information from COCO captions, crafting detailed yet accurate grounded captions while preserving the original referring expressions. This ensures zero error in matching phrases with their corresponding segmentation masks. This technique yields approximately 24K GCG samples. For PSG, we leverage the dataset’s triplet structures, which describe relations between two objects in a scene. These triplets are integrated with COCO captions using GPT-4, resulting in densely annotated captions that can be mapped to segmentation masks. This gives us around 31K additional GCG samples. For Flickr-30K, we use the 158K Flickr captions and their referring expressions alongside associated bounding boxes. These boxes are then accurately segmented using HQ-SAM [17]. In addition, we contribute a minor, high-quality manual annotation set to benchmark the GCG task. Using GranD’s automatic annotations as a base, annotators refine referring expressions to match SAM GT masks, yielding around 1000 focused samples for training and 1000 for evaluation (refer to Appendix D and Fig. 14 for designed prompts and dataset visualizations). This paper is available on arxiv under CC BY 4.0 DEED license. Authors: (1) Hanoona Rasheed, Mohamed bin Zayed University of AI and equally contributing first authors; (2) Muhammad Maaz, Mohamed bin Zayed University of AI and equally contributing first authors; (3) Sahal Shaji, Mohamed bin Zayed University of AI; (4) Abdelrahman Shaker, Mohamed bin Zayed University of AI; (5) Salman Khan, Mohamed bin Zayed University of AI and Australian National University; (6) Hisham Cholakkal, Mohamed bin Zayed University of AI; (7) Rao M. Anwer, Mohamed bin Zayed University of AI and Aalto University; (8) Eric Xing, Mohamed bin Zayed University of AI and Carnegie Mellon University; (9) Ming-Hsuan Yang, University of California - Merced and Google Research; (10) Fahad S. Khan, Mohamed bin Zayed University of AI and Linköping University. Authors: Authors: (1) Hanoona Rasheed, Mohamed bin Zayed University of AI and equally contributing first authors; (2) Muhammad Maaz, Mohamed bin Zayed University of AI and equally contributing first authors; (3) Sahal Shaji, Mohamed bin Zayed University of AI; (4) Abdelrahman Shaker, Mohamed bin Zayed University of AI; (5) Salman Khan, Mohamed bin Zayed University of AI and Australian National University; (6) Hisham Cholakkal, Mohamed bin Zayed University of AI; (7) Rao M. Anwer, Mohamed bin Zayed University of AI and Aalto University; (8) Eric Xing, Mohamed bin Zayed University of AI and Carnegie Mellon University; (9) Ming-Hsuan Yang, University of California - Merced and Google Research; (10) Fahad S. Khan, Mohamed bin Zayed University of AI and Linköping University. Editor's Note: This is Part 4 of 10 of a study detailing the development of an AI model that is designed to describe images to users. Read the rest below. Editor's Note: This is Part 4 of 10 of a study detailing the development of an AI model that is designed to describe images to users. Read the rest below. Editor's Note: This is Part 4 of 10 of a study detailing the development of an AI model that is designed to describe images to users. Read the rest below. Editor's Note: This is Part 4 of 10 of a study detailing the development of an AI model that is designed to describe images to users. Read the rest below. Table of Links Abstract and 1 Introduction 2. Related Work 3. Method 4. Data Annotation Pipeline 5. Experiments 6. Conclusion and References Abstract and 1 Introduction Abstract and 1 Introduction 2. Related Work 2. Related Work 3. Method 3. Method 4. Data Annotation Pipeline 4. Data Annotation Pipeline 5. Experiments 5. Experiments 6. Conclusion and References 6. Conclusion and References Supplementary Material (Part 1) Supplementary Material (Part 1) A. Additional Implementation Details B. Additional Downstream Tasks C. Additional Qualitative Results A. Additional Implementation Details A. Additional Implementation Details B. Additional Downstream Tasks B. Additional Downstream Tasks C. Additional Qualitative Results C. Additional Qualitative Results Supplementary Material (Part 2) Supplementary Material (Part 2) D. Dataset Visualization E. Limitations and Future Work F. Ethics and Societal Impact D. Dataset Visualization D. Dataset Visualization E. Limitations and Future Work E. Limitations and Future Work F. Ethics and Societal Impact F. Ethics and Societal Impact 4. Data Annotation Pipeline We introduce our automated annotation pipeline used to create the Grounding-anything Dataset (GranD). GranD is a comprehensive, multi-purpose image-text dataset offering a range of contextual information, from fine-grained to high-level details. It aims to overcome challenges in image understanding and dense pixel-level grounding, thereby expanding the capabilities of visual instruction tuning in LMMs. The pipeline contains four distinct levels (see Fig. 4). i) Level-1 focuses on object localization and provides semantic labels, segmentation masks, attributes, and depth information. ii) Level-2 defines relationships between detected objects. iii) Level-3 organizes information from the first two levels into a hierarchical scene graph, used to generate dense captions using LLM with in-context examples. iv) Level-4 offers enriched contextual information for a deeper understanding of the scene, going beyond what’s observed (e.g., historical information of a landmark). Please refer to Appendix A.4 for pipeline implementation details. 4.1. Object Localization and Attributes (Level-1) In level-1, the focus is on detailed object identification within images. First, object-bounding boxes are identified using multiple SoTA object detection models. Classagnostic NMS is applied to each model to filter out false positives. After this step, bounding boxes from different models are compared using IoU, with a bounding box retained as an object only if detected by at least two other detection models. We also generate attributes for each filtered object using region-based vision-language models and incorporate depth information to contextualize each object’s relative position within the scene. 4.2. Relationships and Landmarks (Level-2) In level-2, multiple short textual descriptions of the overall scene are generated. Phrases extracted from these descriptions are grounded to specific objects in level-1 to form relationships. These relationships articulate connections between multiple objects or define an object’s role within the scene. Further, each scene is assigned a landmark category that includes a primary and a more specific sub-category (see Tab. 7 in Appendix 7). 4.3. Scene Graph and Dense Captioning (Level-3) In level-3, object attributes and labels from level-1 are combined with the relationships and phrases obtained from level-2 to form a hierarchical scene graph. This structured data serves as a query for LLM to generate dense image captions. To provide additional context, depth values and bounding box coordinates are used to assign each object to specific spatial layers within the scene, such as immediate foreground, foreground, midground, or background. Additionally, short scene-level captions are incorporated into the scene graph to enhance LLMs’ contextual understanding. Dense Captioning Verification: To enhance the fidelity of the LLM-generated dense captions, we implement an automatic verification pipeline using chain-of-thoughts prompting. This pipeline produces a checklist of objects derived from the generated dense caption expected to be present in the image. The associated caption is flagged as inaccurate if any object specified in the checklist is absent from the scene graph. Such captions are then regenerated, incorporating feedback from the initial assessment. Dense Captioning Verification: 4.4. Extra Contextual Insights (Level-4) Level-4 builds on the scene graph from level-3 to obtain a more detailed visual understanding. we query LLM to extract extended contextual insights beyond basic object identification and relationships, including details about the landmarks, historical context, guidelines for interacting with the scene, and even predictive elements about future events. To facilitate this, we prompt LLM with in-context examples. Utilizing our automated annotation pipeline, we annotate a corpus of 11M SAM images [18], which are inherently diverse, high-resolution, and privacy-compliant. The resulting dataset comprises 810M regions, each associated with a segmentation mask, and includes 7.5M unique concepts. Further, the dataset features 84M referring expressions, 22M grounded short captions, and 11M densely grounded captions. To our knowledge, this is the first dataset of this scale generated entirely through an automated annotation pipeline (see Tab. 2 for details and Fig. 15 for dataset sample visualizations). We extend open-source datasets—namely Flickr30K [37], RefCOCOg [16], and PSG [49] by generating compatible GCG annotations. For RefCOCOg, we use the dataset’s referring expressions and their connected masks. These expressions offer concise descriptions of distinct objects in the image. With the aid of GPT-4, we seamlessly blend these referring expressions with contextual information from COCO captions, crafting detailed yet accurate grounded captions while preserving the original referring expressions. This ensures zero error in matching phrases with their corresponding segmentation masks. This technique yields approximately 24K GCG samples. For PSG, we leverage the dataset’s triplet structures, which describe relations between two objects in a scene. These triplets are integrated with COCO captions using GPT-4, resulting in densely annotated captions that can be mapped to segmentation masks. This gives us around 31K additional GCG samples. For Flickr-30K, we use the 158K Flickr captions and their referring expressions alongside associated bounding boxes. These boxes are then accurately segmented using HQ-SAM [17]. In addition, we contribute a minor, high-quality manual annotation set to benchmark the GCG task. Using GranD’s automatic annotations as a base, annotators refine referring expressions to match SAM GT masks, yielding around 1000 focused samples for training and 1000 for evaluation (refer to Appendix D and Fig. 14 for designed prompts and dataset visualizations). This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv

Researchers Develop Groundbreaking Method to Teach AI to 'Understand' Images Like Never Before

Learn how to programmatically fetch and embed your liked Bluesky posts using authentication, API calls, and oEmbed endpoints. I've recently made the move over to Bluesky. I can already confirm that there is a vibrant tech community there with tons of interesting, useful, and inspiring content. I'm a happy new user! As a result, I've been wanting to embed my top liked Bluesky posts in my "Dev roundup" monthly newsletter posts. My aim is to provide a curated list of Bluesky posts that is specifically tailored to Software Developers. Luckily, Bluesky's API is completely free to use, allowing programmatic access to all of the content within. This tutorial will walk you through the process of retrieving and embedding liked Bluesky posts using their API, perfect for personal blogs, portfolios, or content aggregation projects. Understanding my Bluesky API Workflow I have built a script that allows me to automatically embed my Bluesky posts in a markdown blog post. I think that any or all of the steps used in this script are valuable for many use-cases. To summarize my workflow for embedding liked posts, we follow these key steps: Create an authenticated session
Retrieve liked post URIs for an "actor"
Use these URIs to fetch oEmbed embed HTML
Clean and format the embed code The Complete Implementation Let's break down each function and its purpose: 1. Creating a Bluesky Session export const createSession = async (): Promise<string | null> => {
  try {
    const response = await fetch(
      "https://bsky.social/xrpc/com.atproto.server.createSession",
      {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
        },
        body: JSON.stringify({
          identifier: "your-handle",
          password: "your-password",
        }),
      }
    );
    const responseJson = await response.json();
    return responseJson.accessJwt;
  } catch (e) {
    console.error("Failed to create session: ", e);
    return null;
  }
}; Key Insights: This function authenticates your Bluesky account.

Note: this example hardcodes strings for credentials, but this should be avoided in production use cases.


It returns an accessJwt JWT (JSON Web Token) for subsequent API calls
Uses the createSession endpoint from Bluesky's ATP (Authenticated Transfer Protocol)
Error handling ensures graceful failure if authentication fails 2. Retrieving Liked Post URIs export const getBlueskyLikeUris = async (actor: string, limit: number = 40) => {
  const token = await createSession();
  if (!token) {
    console.error("Failed to get token");
    return;
  }
  
  const response = await fetch(
    "https://bsky.social/xrpc/app.bsky.feed.getActorLikes?actor=${actor}&limit=${limit}",
    {
      method: "GET",
      headers: {
        Authorization: `Bearer ${token}`,
      },
    }
  );
  
  const responseJson = await response.json();
  const uris = responseJson.feed.map((entry: any) => entry.post.uri);

  return uris;
}; Key Insights: Requires an authenticated session token
Uses the getActorLikes endpoint to fetch liked posts
Important -- the endpoint domain should be https://bsky.social, as this is an authenticated request.
Extracts unique URIs for each liked post
Limits to 40 posts (configurable) 3. Converting URIs to Embeddable HTML export const getBlueskyPostEmbedMarkup = async (uri: string) => {
  try {
    const response = await fetch(`https://embed.bsky.app/oembed?url=${uri}`);
    const responseJson = await response.json();
    
    const formattedHTML = prettier.format(responseJson.html, {
      parser: "html",
      plugins: [require("prettier/parser-html")],
      htmlWhitespaceSensitivity: "ignore",
      printWidth: 1000,
    });
    
    return formattedHTML.replace(/<script[\s\S]*?<\/script>/g, "");
  } catch (e) {
    console.error("Failed to get Bluesky post embed markup");
    return null;
  }
}; Key Insights: Uses Bluesky's oEmbed endpoint with post URIs to access a post's embeddable HTML
Optional: Utilizes prettier to format the HTML consistently
Optional: Removes <script> tags for security and clean embedding

The reason for this is that I embed a single Bluesky script for each post containing Bluesky content.


Flexible error handling Putting It All Together: A Complete Example async function embedLikedPosts() {
  try {
    // Get liked post URIs
    const likedPostUris = await getBlueskyLikeUris();
    
    if (!likedPostUris) {
      console.error("No liked posts found");
      return;
    }
    
    // Convert URIs to embed HTML
    const embedPromises = likedPostUris.map(getBlueskyPostEmbedMarkup);
    const embedHtmlArray = await Promise.all(embedPromises);
    
    // Filter out any failed embeds
    const validEmbeds = embedHtmlArray.filter(embed => embed !== null);
    
    // Return the markup for all liked posts
    return `
      ## Some Fave Posts 🦋
      ${validEmbeds.join(`\n\n`)}
    `
  } catch (error) {
    console.error("Error embedding Bluesky posts:", error);
  }
} Potential Enhancements This solution works for me because all that I require is a statically generated monthly blog post. Some improvements could include: Add pagination support for fetching more than 40 liked posts
Implement caching to reduce unnecessary API calls
Create a more robust error handling mechanism
Creating mechanism for refreshing accessJwt token if used in long running processes
Sorting liked posts by popularity (likes) Troubleshooting Tips Verify your Bluesky credentials are correct
Check that the Bearer token is being set correctly on your authenticated requests.
Verify that the endpoint domains you are using are all valid. Conclusion Embedding Bluesky posts provides a dynamic way to showcase your social media interactions. By understanding the API workflow and implementing robust error handling, you can create engaging, personalized, and curated content integrations. Next Steps Experiment with the code
Customize the embed styling
Explore additional Bluesky API endpoints Enjoy and happy tinkering! 🚀 Learn how to programmatically fetch and embed your liked Bluesky posts using authentication, API calls, and oEmbed endpoints. Learn how to programmatically fetch and embed your liked Bluesky posts using authentication, API calls, and oEmbed endpoints. I've recently made the move over to Bluesky . I can already confirm that there is a vibrant tech community there with tons of interesting, useful, and inspiring content. I'm a happy new user! As a result, I've been wanting to embed my top liked Bluesky posts in my "Dev roundup" monthly newsletter posts. My aim is to provide a curated list of Bluesky posts that is specifically tailored to Software Developers. Bluesky Luckily, Bluesky's API is completely free to use, allowing programmatic access to all of the content within. This tutorial will walk you through the process of retrieving and embedding liked Bluesky posts using their API, perfect for personal blogs, portfolios, or content aggregation projects. Bluesky's API Understanding my Bluesky API Workflow I have built a script that allows me to automatically embed my Bluesky posts in a markdown blog post. I think that any or all of the steps used in this script are valuable for many use-cases. To summarize my workflow for embedding liked posts, we follow these key steps: Create an authenticated session Retrieve liked post URIs for an "actor" Use these URIs to fetch oEmbed embed HTML Clean and format the embed code Create an authenticated session Retrieve liked post URIs for an "actor" Use these URIs to fetch oEmbed embed HTML oEmbed embed HTML Clean and format the embed code The Complete Implementation Let's break down each function and its purpose: 1. Creating a Bluesky Session export const createSession = async (): Promise<string | null> => {
  try {
    const response = await fetch(
      "https://bsky.social/xrpc/com.atproto.server.createSession",
      {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
        },
        body: JSON.stringify({
          identifier: "your-handle",
          password: "your-password",
        }),
      }
    );
    const responseJson = await response.json();
    return responseJson.accessJwt;
  } catch (e) {
    console.error("Failed to create session: ", e);
    return null;
  }
}; export const createSession = async (): Promise<string | null> => {
  try {
    const response = await fetch(
      "https://bsky.social/xrpc/com.atproto.server.createSession",
      {
        method: "POST",
        headers: {
          "Content-Type": "application/json",
        },
        body: JSON.stringify({
          identifier: "your-handle",
          password: "your-password",
        }),
      }
    );
    const responseJson = await response.json();
    return responseJson.accessJwt;
  } catch (e) {
    console.error("Failed to create session: ", e);
    return null;
  }
}; Key Insights: Key Insights: This function authenticates your Bluesky account.

Note: this example hardcodes strings for credentials, but this should be avoided in production use cases. It returns an accessJwt JWT (JSON Web Token) for subsequent API calls Uses the createSession endpoint from Bluesky's ATP (Authenticated Transfer Protocol) Error handling ensures graceful failure if authentication fails This function authenticates your Bluesky account. Note: this example hardcodes strings for credentials, but this should be avoided in production use cases. Note: this example hardcodes strings for credentials, but this should be avoided in production use cases. Note: this example hardcodes strings for credentials, but this should be avoided in production use cases. It returns an accessJwt JWT (JSON Web Token) for subsequent API calls accessJwt Uses the createSession endpoint from Bluesky's ATP (Authenticated Transfer Protocol) createSession Error handling ensures graceful failure if authentication fails 2. Retrieving Liked Post URIs export const getBlueskyLikeUris = async (actor: string, limit: number = 40) => {
  const token = await createSession();
  if (!token) {
    console.error("Failed to get token");
    return;
  }
  
  const response = await fetch(
    "https://bsky.social/xrpc/app.bsky.feed.getActorLikes?actor=${actor}&limit=${limit}",
    {
      method: "GET",
      headers: {
        Authorization: `Bearer ${token}`,
      },
    }
  );
  
  const responseJson = await response.json();
  const uris = responseJson.feed.map((entry: any) => entry.post.uri);

  return uris;
}; export const getBlueskyLikeUris = async (actor: string, limit: number = 40) => {
  const token = await createSession();
  if (!token) {
    console.error("Failed to get token");
    return;
  }
  
  const response = await fetch(
    "https://bsky.social/xrpc/app.bsky.feed.getActorLikes?actor=${actor}&limit=${limit}",
    {
      method: "GET",
      headers: {
        Authorization: `Bearer ${token}`,
      },
    }
  );
  
  const responseJson = await response.json();
  const uris = responseJson.feed.map((entry: any) => entry.post.uri);

  return uris;
}; Key Insights: Key Insights: Requires an authenticated session token Uses the getActorLikes endpoint to fetch liked posts Important -- the endpoint domain should be https://bsky.social, as this is an authenticated request. Extracts unique URIs for each liked post Limits to 40 posts (configurable) Requires an authenticated session token Uses the getActorLikes endpoint to fetch liked posts the getActorLikes Important -- the endpoint domain should be https://bsky.social, as this is an authenticated request. an authenticated request. Extracts unique URIs for each liked post Limits to 40 posts (configurable) 3. Converting URIs to Embeddable HTML export const getBlueskyPostEmbedMarkup = async (uri: string) => {
  try {
    const response = await fetch(`https://embed.bsky.app/oembed?url=${uri}`);
    const responseJson = await response.json();
    
    const formattedHTML = prettier.format(responseJson.html, {
      parser: "html",
      plugins: [require("prettier/parser-html")],
      htmlWhitespaceSensitivity: "ignore",
      printWidth: 1000,
    });
    
    return formattedHTML.replace(/<script[\s\S]*?<\/script>/g, "");
  } catch (e) {
    console.error("Failed to get Bluesky post embed markup");
    return null;
  }
}; export const getBlueskyPostEmbedMarkup = async (uri: string) => {
  try {
    const response = await fetch(`https://embed.bsky.app/oembed?url=${uri}`);
    const responseJson = await response.json();
    
    const formattedHTML = prettier.format(responseJson.html, {
      parser: "html",
      plugins: [require("prettier/parser-html")],
      htmlWhitespaceSensitivity: "ignore",
      printWidth: 1000,
    });
    
    return formattedHTML.replace(/<script[\s\S]*?<\/script>/g, "");
  } catch (e) {
    console.error("Failed to get Bluesky post embed markup");
    return null;
  }
}; Key Insights: Key Insights: Uses Bluesky's oEmbed endpoint with post URIs to access a post's embeddable HTML Optional: Utilizes prettier to format the HTML consistently Optional: Removes <script> tags for security and clean embedding

The reason for this is that I embed a single Bluesky script for each post containing Bluesky content. Flexible error handling Uses Bluesky's oEmbed endpoint with post URIs to access a post's embeddable HTML Bluesky's oEmbed endpoint Optional: Utilizes prettier to format the HTML consistently prettier Optional: Removes <script> tags for security and clean embedding The reason for this is that I embed a single Bluesky script for each post containing Bluesky content. <script> The reason for this is that I embed a single Bluesky script for each post containing Bluesky content. The reason for this is that I embed a single Bluesky script for each post containing Bluesky content. Flexible error handling Putting It All Together: A Complete Example async function embedLikedPosts() {
  try {
    // Get liked post URIs
    const likedPostUris = await getBlueskyLikeUris();
    
    if (!likedPostUris) {
      console.error("No liked posts found");
      return;
    }
    
    // Convert URIs to embed HTML
    const embedPromises = likedPostUris.map(getBlueskyPostEmbedMarkup);
    const embedHtmlArray = await Promise.all(embedPromises);
    
    // Filter out any failed embeds
    const validEmbeds = embedHtmlArray.filter(embed => embed !== null);
    
    // Return the markup for all liked posts
    return `
      ## Some Fave Posts 🦋
      ${validEmbeds.join(`\n\n`)}
    `
  } catch (error) {
    console.error("Error embedding Bluesky posts:", error);
  }
} async function embedLikedPosts() {
  try {
    // Get liked post URIs
    const likedPostUris = await getBlueskyLikeUris();
    
    if (!likedPostUris) {
      console.error("No liked posts found");
      return;
    }
    
    // Convert URIs to embed HTML
    const embedPromises = likedPostUris.map(getBlueskyPostEmbedMarkup);
    const embedHtmlArray = await Promise.all(embedPromises);
    
    // Filter out any failed embeds
    const validEmbeds = embedHtmlArray.filter(embed => embed !== null);
    
    // Return the markup for all liked posts
    return `
      ## Some Fave Posts 🦋
      ${validEmbeds.join(`\n\n`)}
    `
  } catch (error) {
    console.error("Error embedding Bluesky posts:", error);
  }
} Potential Enhancements This solution works for me because all that I require is a statically generated monthly blog post. Some improvements could include: Add pagination support for fetching more than 40 liked posts Implement caching to reduce unnecessary API calls Create a more robust error handling mechanism Creating mechanism for refreshing accessJwt token if used in long running processes Sorting liked posts by popularity (likes) Add pagination support for fetching more than 40 liked posts Implement caching to reduce unnecessary API calls Create a more robust error handling mechanism Creating mechanism for refreshing accessJwt token if used in long running processes accessJwt Sorting liked posts by popularity (likes) Troubleshooting Tips Verify your Bluesky credentials are correct Check that the Bearer token is being set correctly on your authenticated requests. Verify that the endpoint domains you are using are all valid. Verify your Bluesky credentials are correct Check that the Bearer token is being set correctly on your authenticated requests. Verify that the endpoint domains you are using are all valid. Conclusion Embedding Bluesky posts provides a dynamic way to showcase your social media interactions. By understanding the API workflow and implementing robust error handling, you can create engaging, personalized, and curated content integrations. Next Steps Experiment with the code Customize the embed styling Explore additional Bluesky API endpoints Experiment with the code Customize the embed styling Explore additional Bluesky API endpoints Enjoy and happy tinkering! 🚀

Bluesky's Free API Helps Me Curate My Top Posts Like Never Before

Bluesky ukan API Gratis ukax Nayan Jach’a Posts ukanakar Curar ukham yanapt’itu, kunjamatix Janipuniw Nayrjax ukhama

La API gratuita de Bluesky me ayuda a organizar mis publicaciones más destacadas como nunca antes

API رایگان Bluesky به من کمک می‌کند تا پست‌های برتر خود را مانند قبل تنظیم کنم

A Bluesky ingyenes API-ja úgy segít összeválogatni a legnépszerűbb bejegyzéseimet, mint még soha

API Gratis Bluesky Membantu Saya Menyusun Postingan Terbaik Saya Seperti Sebelumnya

Bluesky の無料 API は、これまでにないほど私の人気投稿をキュレートするのに役立ちます

API ya ofele ya Bluesky Esalisaka Ngai Na Curer Ba Posts Na Ngai Ya Likolo Neti Jamais Avant

Bluesky को नि: शुल्क API ले मलाई मेरो शीर्ष पोष्टहरू पहिले कहिल्यै क्युरेट गर्न मद्दत गर्दछ

Bezpłatne API Bluesky pomaga mi w jak nigdy dotąd selekcjonowaniu moich najlepszych postów

د بلوسکي وړیا API زما سره زما د غوره پوسټونو په ترتیب کولو کې مرسته کوي لکه مخکې هیڅکله

Bluesky's API Bilaashka ah waxay iga caawisaa inaan hagaajiyo qoraaladayda ugu sarreeya si aan waligeed horay loo arag

Бесплатни АПИ Блуески-а ми помаже да бирам моје најпопуларније постове као никада раније

Table of Links Abstract and Introduction
Related Work
Problem Definition
Method
Experiments
Conclusion and References A. Appendix A.1. Full Prompts and A.2 ICPL Details A. 3 Baseline Details A.4 Environment Details A.5 Proxy Human Preference A.6 Human-in-the-Loop Preference c Authors:
(1) Chao Yu, Tsinghua University;
(2) Hong Lu, Tsinghua University;
(3) Jiaxuan Gao, Tsinghua University;
(4) Qixin Tan, Tsinghua University;
(5) Xinting Yang, Tsinghua University;
(6) Yu Wang, with equal advising from Tsinghua University;
(7) Yi Wu, with equal advising from Tsinghua University and the Shanghai Qi Zhi Institute;
(8) Eugene Vinitsky, with equal advising from New York University (zoeyuchao@gmail.com). This paper is available on arxiv under CC 4.0 license. Table of Links Abstract and Introduction Related Work Problem Definition Method Experiments Conclusion and References Abstract and Introduction Abstract and Introduction Related Work Related Work Problem Definition Problem Definition Method Method Experiments Experiments Conclusion and References Conclusion and References A. Appendix A. Appendix A.1. Full Prompts and A.2 ICPL Details A.1. Full Prompts and A.2 ICPL Details A. 3 Baseline Details A. 3 Baseline Details A.4 Environment Details A.4 Environment Details A.5 Proxy Human Preference A.5 Proxy Human Preference A.6 Human-in-the-Loop Preference A.6 Human-in-the-Loop Preference c Authors: (1) Chao Yu, Tsinghua University; (2) Hong Lu, Tsinghua University; (3) Jiaxuan Gao, Tsinghua University; (4) Qixin Tan, Tsinghua University; (5) Xinting Yang, Tsinghua University; (6) Yu Wang, with equal advising from Tsinghua University; (7) Yi Wu, with equal advising from Tsinghua University and the Shanghai Qi Zhi Institute; (8) Eugene Vinitsky, with equal advising from New York University (zoeyuchao@gmail.com). Authors: Authors: (1) Chao Yu, Tsinghua University; (2) Hong Lu, Tsinghua University; (3) Jiaxuan Gao, Tsinghua University; (4) Qixin Tan, Tsinghua University; (5) Xinting Yang, Tsinghua University; (6) Yu Wang, with equal advising from Tsinghua University; (7) Yi Wu, with equal advising from Tsinghua University and the Shanghai Qi Zhi Institute; (8) Eugene Vinitsky, with equal advising from New York University ( zoeyuchao@gmail.com ). zoeyuchao@gmail.com This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv

Human Preferences Help Scientists Train AI 30x Faster Than Before

There is a popular saying that: Data is the new gold, and the web (internet) is the biggest gold mine. As a business owner, student, drop shipper, or corporate worker, you must have faced the risk of going out of business if you lack access to some (if not the most) crucial piece of data. But fear not; there is a solution that helps you unlock the full potential of web data scraping/collection, which I will show you in this article. Why Do We Collect Data: The Essence of Data Scraping/Collection Before I dive into the challenges faced with data scraping/collection and the wonders of Bright Data solutions, it is important to understand the essence of data scraping in this digital age that we are in. Data scraping is a powerful technique that efficiently navigates the internet to extract specific valuable informations, acting like a personal assistant for data retrieval. Whether you seek to analyze market trends and performance, track competitor prices, or simply satisfy your thirst for knowledge, data scraping is the key to making your life easier. Challenges With Data Scraping: The Limitations Before Bright Data Solutions Like an adventure, data scraping comes with its share of challenges and obstacles. One significant hurdle is the risk of websites feeling under attack when excessive data is scraped too quickly. To counter such attempts, websites deploy defensive measures, such as CAPTCHAs, to determine whether the user is a human or a computer. If the volume of requests persists, the website may resort to blocking users, hindering or halting data scraping altogether. To avoid such unfavorable situations, a more versatile and intelligent solution is required. Bright Data: The Ultimate Solution for Web Data Collection Bright Data comes as the ultimate solution for web data collection tailored specifically for businesses. With its automated online data collection platform, Bright Data offers unlimited access to the vast wisdom of the web without exposing you to any risks. How Bright Data Works The magic behind Bright Data lies in its web unlocker: a unique P2P network composed of millions of genuine residential and mobile IPs from across the globe. These IPs serve as secret identities, making your data requests appear as if they come from real people, making it highly unlikely for websites to block them. Why I Am Using Bright Data: The Power of Using Bright Data 1. Versatility and Ease of Use Bright Data shines with its user-friendly approach. To start scraping vast amounts of data, all you need is a scraping browser — an automated browser that allows you to interact with websites just like a human user. It efficiently handles challenges such as CAPTCHAs, fingerprints, and retries while mimicking the behavior of a regular user. Compatible with popular automation tools like Puppeteer and Playwright and programming languages like Python and Node.js, the scraping browser makes data collection seamless and efficient. 2. Ready-to-Use Data Sets If you ever need ready-to-use data, Bright Data’s data sets tool provides pre-collected and structured data from various domains, including popular platforms like LinkedIn, Amazon, Crunchbase, Indeed, and Twitter. Regularly updated, these comprehensive data sets offer you a treasure trove of information at your fingertips. Practical Examples of What I Have Used Bright Data for 1. E-Commerce: Monitoring Competitor Prices and Enriching Product Catalogs In the world of e-commerce, web data plays a crucial role in monitoring competitor prices, ensuring product availability, and understanding customer reviews. For instance, a global Beauty and Personal Care Marketplace efficiently uses web data to check competitor prices in real-time, enabling them to adjust pricing strategies and optimize promotions on the fly, enhancing their market competitiveness. 2. Social Media: Analyzing Consumer Sentiment and Trends Social media platforms offer a gold mine of user-generated content, providing valuable insights into consumer behavior, trends, and sentiment. A global digital marketing agency effectively analyzes consumer sentiment towards their client’s brands using web data from various social media platforms. This deeper understanding empowers them to refine marketing strategies and achieve better results. In Conclusion: The Bright Future of Data Collection As I conclude with Bright Data's powerful web data collection solutions, one thing becomes very clear; data scraping is a game-changer in today’s digital landscape. Whether you are a business owner, entrepreneur, or data professional, harnessing the power of web data can unlock countless opportunities for growth and success. Traditional manual data collection methods are slow, labor-intensive, and error-prone, rendering them ineffective and inefficient in the face of today’s data-driven world. By automating the process with Bright Data, you can extract specific and relevant information from web pages effortlessly, propelling your business to new heights. So, if you find yourself in need of vast amounts of data, fast and easy, look no further than Bright Data. Embrace the power of web data, and let it guide you to new horizons of success in this ever-evolving digital landscape. There is a popular saying that: Data is the new gold, and the web (internet) is the biggest gold mine. Data is the new gold, and the web (internet) is the biggest gold mine. Data is the new gold, and the web (internet) is the biggest gold mine. As a business owner, student, drop shipper, or corporate worker, you must have faced the risk of going out of business if you lack access to some (if not the most) crucial piece of data. But fear not; there is a solution that helps you unlock the full potential of web data scraping/collection, which I will show you in this article. (if not the most) Why Do We Collect Data: The Essence of Data Scraping/Collection Before I dive into the challenges faced with data scraping/collection and the wonders of Bright Data solutions, it is important to understand the essence of data scraping in this digital age that we are in. Bright Data Data scraping is a powerful technique that efficiently navigates the internet to extract specific valuable informations, acting like a personal assistant for data retrieval. Data scraping is a powerful technique that efficiently navigates the internet to extract specific valuable informations, acting like a personal assistant for data retrieval. Whether you seek to analyze market trends and performance, track competitor prices, or simply satisfy your thirst for knowledge, data scraping is the key to making your life easier. Challenges With Data Scraping: The Limitations Before Bright Data Solutions Bright Data Like an adventure, data scraping comes with its share of challenges and obstacles. One significant hurdle is the risk of websites feeling under attack when excessive data is scraped too quickly. To counter such attempts, websites deploy defensive measures, such as CAPTCHAs, to determine whether the user is a human or a computer. If the volume of requests persists, the website may resort to blocking users, hindering or halting data scraping altogether. To avoid such unfavorable situations, a more versatile and intelligent solution is required. Bright Data: The Ultimate Solution for Web Data Collection Bright Data comes as the ultimate solution for web data collection tailored specifically for businesses. With its automated online data collection platform, Bright Data offers unlimited access to the vast wisdom of the web without exposing you to any risks. Bright Data How Bright Data Works How Bright Data Works The magic behind Bright Data lies in its web unlocker: a unique P2P network composed of millions of genuine residential and mobile IPs from across the globe. These IPs serve as secret identities, making your data requests appear as if they come from real people, making it highly unlikely for websites to block them. Why I Am Using Bright Data: The Power of Using Bright Data 1. Versatility and Ease of Use 1. Versatility and Ease of Use Bright Data shines with its user-friendly approach. To start scraping vast amounts of data, all you need is a scraping browser — an automated browser that allows you to interact with websites just like a human user. It efficiently handles challenges such as CAPTCHAs, fingerprints, and retries while mimicking the behavior of a regular user. Compatible with popular automation tools like Puppeteer and Playwright and programming languages like Python and Node.js, the scraping browser makes data collection seamless and efficient. 2. Ready-to-Use Data Sets 2. Ready-to-Use Data Sets If you ever need ready-to-use data, Bright Data’s data sets tool provides pre-collected and structured data from various domains, including popular platforms like LinkedIn, Amazon, Crunchbase, Indeed, and Twitter. Regularly updated, these comprehensive data sets offer you a treasure trove of information at your fingertips. Practical Examples of What I Have Used Bright Data for 1. E-Commerce: Monitoring Competitor Prices and Enriching Product Catalogs 1. E-Commerce: Monitoring Competitor Prices and Enriching Product Catalogs In the world of e-commerce, web data plays a crucial role in monitoring competitor prices, ensuring product availability, and understanding customer reviews. For instance, a global Beauty and Personal Care Marketplace efficiently uses web data to check competitor prices in real-time, enabling them to adjust pricing strategies and optimize promotions on the fly, enhancing their market competitiveness. 2. Social Media: Analyzing Consumer Sentiment and Trends 2. Social Media: Analyzing Consumer Sentiment and Trends Social media platforms offer a gold mine of user-generated content, providing valuable insights into consumer behavior, trends, and sentiment. A global digital marketing agency effectively analyzes consumer sentiment towards their client’s brands using web data from various social media platforms. This deeper understanding empowers them to refine marketing strategies and achieve better results. In Conclusion: The Bright Future of Data Collection As I conclude with Bright Data's powerful web data collection solutions, one thing becomes very clear; data scraping is a game-changer in today’s digital landscape. Whether you are a business owner, entrepreneur, or data professional, harnessing the power of web data can unlock countless opportunities for growth and success. Traditional manual data collection methods are slow, labor-intensive, and error-prone, rendering them ineffective and inefficient in the face of today’s data-driven world. By automating the process with Bright Data, you can extract specific and relevant information from web pages effortlessly, propelling your business to new heights. So, if you find yourself in need of vast amounts of data, fast and easy, look no further than Bright Data. Embrace the power of web data, and let it guide you to new horizons of success in this ever-evolving digital landscape.

How Bright Data AI Made Web Data Scraping/Collection Effortless: The Challenges Before Bright Data AI Solutions

Як AI Bright Data зрабіў скрабаванне/збор вэб-даных простым: праблемы перад рашэннямі AI Bright Data

Com Bright Data AI va fer que el raspat/recollida de dades web sigui sense esforç: els reptes abans de les solucions Bright Data AI

Cómo Bright Data AI hizo que la recopilación y el raspado de datos web fueran sencillos: los desafíos que enfrentaban las soluciones de Bright Data AI

Kako je Bright Data AI skrapiranje/prikupljanje web podataka učinio jednostavnim: Izazovi pred Bright Data AI rješenjima

Bright Data AI が Web データのスクレイピング/収集をいかに簡単にしたか: Bright Data AI ソリューションの課題

Cara AI Data Cerah Menjadikan Pengikisan/Pengumpulan Data Web Mudah: Cabaran Sebelum Penyelesaian AI Data Cerah

څنګه روښانه ډیټا AI د ویب ډیټا سکریپینګ/راټولول بې ځایه کړي: د روښانه ډیټا AI حلونو دمخه ننګونې

Bright Data AI විසින් Web Data Scraping/එකතු කිරීම වෙහෙසකින් තොරව සිදු කළේ කෙසේද: Bright Data AI විසඳුම් ඉදිරියේ ඇති අභියෝග

Jinsi Data Angavu AI Ilivyofanya Uchakachuaji/Ukusanyaji wa Data ya Wavuti Bila Jitihada: Changamoto Kabla ya Suluhisho Mkali za Data AI

எப்படி பிரைட் டேட்டா AI இணையத் தரவை ஸ்கிராப்பிங்/சேகரிப்பதை சிரமமின்றி உருவாக்கியது: பிரைட் டேட்டா AI தீர்வுகளுக்கு முன் உள்ள சவால்கள்

کس طرح برائٹ ڈیٹا اے آئی نے ویب ڈیٹا کو اسکریپنگ/جمع کرنے کو آسان بنایا: برائٹ ڈیٹا اے آئی کے حل کے سامنے چیلنجز

Bright Data AI đã giúp việc thu thập/sao chép dữ liệu web trở nên dễ dàng như thế nào: Những thách thức trước khi có giải pháp Bright Data AI

BEFORE

before-inc.com

-- employees

Since n.d.

COMPANY RANKING

BEFORE

EVERGREEN INDEX #923

Representing this company? Click here to claim and customize this page!

Before WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!

BEFORE

before-inc.com

-- employees

Since n.d.

COMPANY RANKING

BEFORE

EVERGREEN INDEX #923

RELATED COMPANIESRANK

STORIES ABOUT BEFORE

Representing this company? Click here to claim and customize this page!

Before WIKI

Apply to get your Tech Company News Page live on HackerNoon.com today!