frag.jetzt issues
https://git.thm.de/arsnova/frag.jetzt/-/issues
2024-01-25T19:55:55+01:00
https://git.thm.de/arsnova/frag.jetzt/-/issues/1459
Restructure and improve code & remove console logs 😅
2024-01-25T19:55:55+01:00
Yousif Wardah
Restructure and improve code & remove console logs 😅
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1458
Token consumption should not display any $ amount if value is set to 0
2024-01-25T19:54:45+01:00
Yousif Wardah
Token consumption should not display any $ amount if value is set to 0
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1457
Stepper requires validators
2024-01-25T19:54:14+01:00
Yousif Wardah
Stepper requires validators
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1456
Stepper requires proper i18n errors
2024-01-25T19:54:05+01:00
Yousif Wardah
Stepper requires proper i18n errors
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1453
Prompt Compression
2024-02-28T08:04:29+01:00
Klaus-Dieter Quibeldey-Cirkel
Prompt Compression
Try to implement:
GitHub: https://github.com/microsoft/LLMLingua
See:
- https://llmlingua.com/
- Paper: https://arxiv.org/pdf/2310.05736.pdf
> The paper titled "LLMLingua: Compressing Prompts for Accelerated Inference of Large Languag...
Try to implement:
GitHub: https://github.com/microsoft/LLMLingua
See:
- https://llmlingua.com/
- Paper: https://arxiv.org/pdf/2310.05736.pdf
> The paper titled "LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models" introduces a method for prompt compression in large language models (LLMs). The method, named LLMLingua, is designed to address the challenge of increasingly lengthy prompts in LLMs, which can exceed tens of thousands of tokens. LLMLingua uses a coarse-to-fine prompt compression approach, featuring a budget controller for maintaining semantic integrity at high compression ratios, an iterative token-level compression algorithm for modeling interdependencies, and an instruction tuning method for distribution alignment between models. The paper demonstrates the effectiveness of LLMLingua across various datasets, showing that it can achieve significant compression with minimal performance loss. This approach is particularly relevant for accelerating model inference and reducing computational costs in practical applications of LLMs.
- Article: https://www.linkedin.com/pulse/prompt-compression-large-language-models-ashish-bhatia-vt4ye/
Ruben Bimberg
Ruben Bimberg
2024-02-29
https://git.thm.de/arsnova/frag.jetzt/-/issues/1452
Bug: Incorrect formatting of ChatGPT output
2023-12-31T09:14:21+01:00
Klaus-Dieter Quibeldey-Cirkel
Bug: Incorrect formatting of ChatGPT output
From a user:
> Lieber Klaus,
>
> GPT4 ist super - umso besser, dass Du irgendwoher die Lizenz für uns alle organisiert hast.
> Das ist gut und wichtig! Danke!
>
> Schade nur, dass das Frontend nicht so richtig brauchbar ist. Wie soll ...
From a user:
> Lieber Klaus,
>
> GPT4 ist super - umso besser, dass Du irgendwoher die Lizenz für uns alle organisiert hast.
> Das ist gut und wichtig! Danke!
>
> Schade nur, dass das Frontend nicht so richtig brauchbar ist. Wie soll ein Studenty mit so einer Ausgabe etwas anfangen (s. Anhang).
>
> Ich kenne mich mit dem Web-Zeugs nicht so gut aus, wie mit neuronalen Netzen, aber es kann eigentlich nicht viel schwieriger sein, Code zu formatieren, als einen Transformer zu bauen?
>
> Oder habe ich da noch einen Ad-Blocker aktiviert und bin selbst schuld?
>
> Danke für Deine Hilfe
> (a)ndreas
![Screenshot_from_2023-12-30_17-30-54](/uploads/38c3181c57f9037b6f788f4c1a3767e3/Screenshot_from_2023-12-30_17-30-54.png)
Ruben Bimberg
Ruben Bimberg
2024-01-09
https://git.thm.de/arsnova/frag.jetzt/-/issues/1435
Always place the cursor in the first input field
2023-12-21T15:59:43+01:00
Anas Ahakki
Always place the cursor in the first input field
**Title**: Default Cursor Placement in First Input Field<br>
**Description**: Improve user experience by automatically positioning the cursor in the initial input field by default. This enhancement ensures that upon loading the interface...
**Title**: Default Cursor Placement in First Input Field<br>
**Description**: Improve user experience by automatically positioning the cursor in the initial input field by default. This enhancement ensures that upon loading the interface, user focus is directed to the first input field, streamlining data entry and enhancing usability
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1434
Increase the distance between two toggle switches
2023-12-21T15:59:41+01:00
Anas Ahakki
Increase the distance between two toggle switches
**Title**: Increased Spacing Between Toggle Switches<br>
**Description**: Optimize user experience by widening the gap between two adjacent toggle switches. This modification aims to improve visual clarity and interaction ease by providi...
**Title**: Increased Spacing Between Toggle Switches<br>
**Description**: Optimize user experience by widening the gap between two adjacent toggle switches. This modification aims to improve visual clarity and interaction ease by providing ample space between toggle elements, reducing accidental selection and enhancing overall usability within the interface
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1432
Wrap texts on toggle switches
2024-01-22T23:01:19+01:00
Anas Ahakki
Wrap texts on toggle switches
**Title**: Text Wrapping for Toggle Switches<br>
**Description**: Implement text wrapping functionality for toggle switches to accommodate longer labels or descriptions. This enhancement aims to ensure proper display and readability of t...
**Title**: Text Wrapping for Toggle Switches<br>
**Description**: Implement text wrapping functionality for toggle switches to accommodate longer labels or descriptions. This enhancement aims to ensure proper display and readability of text content within toggle switches, providing a better user experience, especially when dealing with longer or multiline text
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1429
Voucher | API-Key
2024-01-16T09:53:39+01:00
Anas Ahakki
Voucher | API-Key
**Title**: Voucher and API Key Integration<br>
**Description**: Implement functionality to incorporate both voucher and API key inputs. This feature enables users to apply vouchers or enter API keys, enhancing system accessibility and en...
**Title**: Voucher and API Key Integration<br>
**Description**: Implement functionality to incorporate both voucher and API key inputs. This feature enables users to apply vouchers or enter API keys, enhancing system accessibility and enabling seamless interactions within the platform
---
**Details**:
1. I have a voucher (default: no)
2. **If yes**: Show input field for voucher
Show status and show voucher usage: "Perfect: You are taking part in the evaluation study. The evaluation study gives you $6.00 per month. You have already used $3.97 this month."
3. **if no**:
- API key input field
- Organisation ID input field
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1428
Step 1: Voucher | API key
2023-12-21T15:37:50+01:00
Anas Ahakki
Step 1: Voucher | API key
ARS-PLE ready for UX tests
https://git.thm.de/arsnova/frag.jetzt/-/issues/1421
Sprint Goal: Enhance User Interaction and Platform Customization
2023-12-19T08:37:07+01:00
Klaus-Dieter Quibeldey-Cirkel
Sprint Goal: Enhance User Interaction and Platform Customization
**Objective:**
To significantly improve the user interaction and customization capabilities of the frag.jetzt platform, making it more dynamic, personalized, and user-friendly while ensuring a secure and intuitive experience across diff...
**Objective:**
To significantly improve the user interaction and customization capabilities of the frag.jetzt platform, making it more dynamic, personalized, and user-friendly while ensuring a secure and intuitive experience across different user roles and functionalities.
**Sprint Deliverables:**
1. **File and Picture Upload Feature:**
- Implement file and picture upload options in the room configuration for chatbot interactions.
- Ensure that participants can share multimedia content to facilitate more dynamic and engaging conversations.
2. **Dynamic Interface Customization:**
- Develop a system for the interface to dynamically adjust options, menus, labels, tooltips, and explanations based on the user's role (educator or learner) and chosen functionality (ARS or PLE).
- Conduct user testing to confirm the intuitive nature of these adjustments.
3. **Token Quota Management Stepper:**
- Create a straightforward, step-by-step process for managing OpenAI API token quotas, designed for users with no prior technical knowledge.
- Integrate clear explanations and guidance within the stepper to assist beginners in managing their API usage.
4. **Room-Specific Prompt Presets Integration:**
- Provide room creators with the ability to consolidate and select prompt presets from general or private prompt catalogs.
- Facilitate a more efficient customization process for room creators to tailor the room experience.
5. **Keycloak Integration Enhancement:**
- Complete the integration of Keycloak to ensure a fully functional and consistent user authentication and management system.
- Enhance security and streamline the login process to provide a seamless user experience.
**Acceptance Criteria:**
- All features must be tested for functionality and ease of use.
- The system must maintain high performance and stability with the new features.
- User feedback should be gathered to ensure the new features meet their needs.
- Documentation should be updated to reflect new features and guide users on how to use them.
- Security assessments should be conducted to ensure that user data is protected, especially with the new file upload feature.
**Testing and Review:**
- Conduct usability testing sessions for each new feature with actual users from both educator and learner demographics.
- Review the security of the file upload feature to ensure that it does not introduce any vulnerabilities.
- Perform a code review and security audit for all new integrations.
By focusing on these key areas, the upcoming sprint aims to deliver a set of features that will enrich user engagement and streamline their experience on the frag.jetzt platform.
https://git.thm.de/arsnova/frag.jetzt/-/issues/1412
MatStepper Component for Room Configuration
2023-12-21T13:54:55+01:00
Klaus-Dieter Quibeldey-Cirkel
MatStepper Component for Room Configuration
**Overview:**
This user story outlines a series of step headings for the `MatStepper` component, designed to streamline and enhance the room configuration process on the "frag.jetzt" platform. Each step is crafted to guide users in an i...
**Overview:**
This user story outlines a series of step headings for the `MatStepper` component, designed to streamline and enhance the room configuration process on the "frag.jetzt" platform. Each step is crafted to guide users in an intuitive and structured manner.
1. **Intro** (Registration and ChatGPT Activation Check)
- The first step involves verifying user registration and ChatGPT activation, ensuring that necessary prerequisites are met for room configuration.
2. **Select Role**
- Users select their role, which tailors the configuration process to their specific needs and access rights.
3. **Define Room Access**
- In this step, users establish the room's identity by setting up the room name and defining an access code, ensuring controlled entry to the room.
4. **Configure Room** (Standard or Custom)
- Users choose between a standard configuration for quick setup or opt for a custom configuration to personalize room settings further.
5. **ChatGPT Options** (If Available)
- This step allows users to customize ChatGPT settings, enhancing the room's interactive capabilities, provided ChatGPT is available.
6. **Customize Features**
- Users configure additional features, grouped into intuitive categories like Q&A/Topic-Board, Interaction, Bonus Archive, and Presentation for streamlined selection.
7. **Review & Confirm**
- A summary of all configurations is presented for review and confirmation, allowing users to ensure all settings are as intended before finalizing.
**Objective:**
The objective is to implement these steps as headings in the `MatStepper` component to improve the room creation feature on the "frag.jetzt" platform, making it more intuitive, user-friendly, and efficient, thereby enhancing the overall user experience.
**User Story:**
As a user of the "frag.jetzt" platform, I want to be able to open a modal by clicking the existing plus icon on the homepage, which guides me through a hierarchical yes/no decision system for room configuration, to make the creation process more streamlined and user-friendly.
**Acceptance Criteria:**
1. **Utilization of the Existing Plus Icon:**
- The plus icon on the homepage is used as the entry point for initiating the room creation process.
2. **Modal for the Decision Process:**
- Clicking the plus icon opens a modal that encapsulates the complete room creation process using the `MatStepper` component.
3. **Intro (Registration and ChatGPT Activation Check):**
- The first step ensures the user is registered and ChatGPT is activated (if applicable).
4. **Role Selection:**
- Users select their role, influencing subsequent configuration options and access rights.
5. **Room Access (Room Name and Access Code):**
- Users set up the room name and define an access code for room entry.
6. **Room Configuration (Standard or Custom):**
- Users choose between a standard or custom configuration, affecting the level of customization for the room setup.
7. **ChatGPT Settings (If Available):**
- If ChatGPT is available, users configure its settings to enable interactive features in the room.
8. **Feature Configuration:**
- Users configure additional features, categorized for ease of selection (e.g., Q&A/Topic-Board, Interaction, Bonus Archive, Presentation).
9. **Summary (Review and Confirmation):**
- Users review and confirm all settings in a summarized view, with the option to go back and modify selections.
10. **Room Creation (Finalization):**
- The final step involves confirming the creation of the room, with a success message or further instructions.
11. **User-Friendly Interface:**
- The modal interface is intuitive, with toggle switches or input fields for decisions, and allows users to navigate back through the steps.
12. **Visualization of Decision Paths:**
- Each step visually indicates the progression and the number of remaining steps in the process.
13. **Quality Assurance:**
- The modal and its decision-making process are thoroughly tested to ensure they are intuitive, user-friendly, and free from errors.
Here's a table presenting the `MatStepper` step titles in English, German, and French:
| Step Number | English | German | French |
|-------------|-----------------------------|---------------------------------|---------------------------------|
| 1 | Intro | Einführung | Intro : |
| 2 | Select Role | Rolle wählen | Sélectionner le Rôle |
| 3 | Define Room Access | Raumzugang festlegen | Définir l'Accès à la Salle |
| 4 | Configure Room | Raum konfigurieren | Configurer la Salle |
| (5) | ChatGPT Options | ChatGPT-Optionen | Options ChatGPT |
| (6) | Customize Features | Features anpassen | Personnaliser les Fonctionnalités |
| (7) | Review & Confirm | Überprüfen & bestätigen | Réviser & Confirmer |
![activity_diagram_room_creation.svg](/uploads/ae2993cb23b62d01bfe0ac1bf8b927ae/activity_diagram_room_creation.svg)
ARS-PLE ready for UX tests
Ruben Bimberg
Ruben Bimberg
2023-12-21
https://git.thm.de/arsnova/frag.jetzt/-/issues/1406
Enhancing Keycloak Integration for Improved User Experience
2023-12-19T11:23:11+01:00
Ruben Bimberg
Enhancing Keycloak Integration for Improved User Experience
**As a** user of the platform,
**I want** the Keycloak integration to be fully completed and consistent,
**So that** I can enjoy a seamless, intuitive, and secure user experience.
# Acceptance Criteria:
- [ ] **Consistent Design**: ...
**As a** user of the platform,
**I want** the Keycloak integration to be fully completed and consistent,
**So that** I can enjoy a seamless, intuitive, and secure user experience.
# Acceptance Criteria:
- [ ] **Consistent Design**: The Account Console should match the platform's overall design for a unified user experience.
- [ ] **Logout Functionality**: There should be a clear and easily accessible logout feature to ensure users can securely end their sessions.
- [ ] **Streamlined Account Deletion**: The process for deleting an account should be straightforward without unnecessary redirections to enhance user friendliness.
- [ ] **Clear User Guidance**: All interfaces and functionalities related to Keycloak should provide clear instructions and support to avoid confusion and frustration for users.
# Description:
Currently, the platform faces several challenges due to the incomplete integration of Keycloak. Users encounter difficulties with the absence of a logout function, raising security concerns. Additionally, the process for account deletion involves numerous redirections, negatively impacting user experience. The differing design in the Account Console also creates confusion and disrupts the seamless experience on the platform.
This User Story aims to address these issues by completing and enhancing the integration of Keycloak. Implementing a consistent design, clear logout functionality, simplified account deletion process, and improved user guidance will create an intuitive and secure user experience. This will increase user satisfaction and trust in the platform.
ARS-PLE ready for UX tests
Ruben Bimberg
Ruben Bimberg
2024-01-18
https://git.thm.de/arsnova/frag.jetzt/-/issues/1405
Navigation Enhancement with a Navigation Bar
2023-11-22T12:09:57+01:00
Anas Ahakki
Navigation Enhancement with a Navigation Bar
## Description
As a user, I want the navigation on the website to be improved by implementing a navigation bar for easier access to different sections.
### Acceptance Criteria:
1. The navigation bar should be prominently displayed at ...
## Description
As a user, I want the navigation on the website to be improved by implementing a navigation bar for easier access to different sections.
### Acceptance Criteria:
1. The navigation bar should be prominently displayed at the top of the website on all pages.
2. The navigation bar should include clear and concise labels for each section or category.
3. The navigation bar should be responsive, adapting to different screen sizes and devices.
4. Users should be able to easily identify the current active section through a visual indicator on the navigation bar.
5. The navigation bar should facilitate quick navigation between main sections and sub-sections.
6. Implement a hover effect or highlight mechanism to enhance the visibility of each navigation item.
7. Ensure that the navigation bar remains consistent across all pages of the website.
8. The navigation bar should support any future additions or modifications to the website's structure.
9. Test the navigation bar's functionality on various browsers to ensure cross-browser compatibility.
10. Evaluate user feedback to further refine and optimize the navigation bar for an improved user experience.
https://git.thm.de/arsnova/frag.jetzt/-/issues/1404
Clear guidance on how to use prompts
2023-11-23T08:08:41+01:00
Donald Donchi Fofack
Clear guidance on how to use prompts
## User Story
**As** a user of the "fragt.jetzt" platform,
**I want** clear guidance on how to use prompts in the chatbot,
**so that** I can more effectively use the chatbot's features and achieve better outcomes.
## Description
Par...
## User Story
**As** a user of the "fragt.jetzt" platform,
**I want** clear guidance on how to use prompts in the chatbot,
**so that** I can more effectively use the chatbot's features and achieve better outcomes.
## Description
Participants in the test showed confusion about the meaning and application of prompts during chatbot interaction. This confusion hampers the user experience and prevents efficient use of the chatbot.
## Acceptance Criteria
1. **Intuitive Prompt Introduction:** A simple and easily understandable introduction to prompts should be displayed during the first chatbot interaction.
2. **Examples and Guidance:** The chatbot interface should provide examples and tips for using prompts.
3. **Feedback Mechanisms:** The chatbot should offer constructive feedback on the prompts entered to facilitate learning.
4. **FAQ or Tutorial Integration:** An FAQ section or an interactive tutorial for effective use of prompts could provide additional support for users.
5. **Adaptive UI Elements:** The user interface should be adaptive, presenting frequently used prompts or personalized suggestions.
https://git.thm.de/arsnova/frag.jetzt/-/issues/1400
Standardization of ChatGPT Activation in Room Settings
2023-11-22T12:17:45+01:00
Klaus-Dieter Quibeldey-Cirkel
Standardization of ChatGPT Activation in Room Settings
**User Story:**
As a room creator, I expect ChatGPT to be a standard, pre-activated feature in my room settings, enabling immediate and seamless interaction with the AI for participants without requiring manual activation on my part.
*...
**User Story:**
As a room creator, I expect ChatGPT to be a standard, pre-activated feature in my room settings, enabling immediate and seamless interaction with the AI for participants without requiring manual activation on my part.
**Acceptance Criteria:**
1. ChatGPT is automatically enabled in all new rooms created on the platform, making it a default feature.
2. Room creators have the option to disable ChatGPT if it is not needed for their specific room, but the default state upon room creation is 'enabled'.
3. The presence of ChatGPT is clearly indicated in the room settings, along with options to customize its behavior and parameters as per the room's requirements.
4. Information regarding the default activation of ChatGPT and how to disable it, if necessary, is prominently displayed or easily accessible to room administrators.
5. Despite being a default feature, the integration of ChatGPT respects all privacy and security standards, ensuring the safety and confidentiality of user interactions.
6. Adequate support and guidelines are provided to educate room creators about the functionalities and benefits of having ChatGPT as a standard feature in their rooms.
https://git.thm.de/arsnova/frag.jetzt/-/issues/1397
Integration of Room-Specific Prompt Presets as Selectable Sets from General o...
2023-12-19T11:21:48+01:00
Klaus-Dieter Quibeldey-Cirkel
Integration of Room-Specific Prompt Presets as Selectable Sets from General or Private Prompt Catalogs
**User Story:**
As a room creator, I want the ability to consolidate various prompt presets of a room into a set, which can be selected from either the general or private prompt catalog, enabling me to streamline the prompt selection pr...
**User Story:**
As a room creator, I want the ability to consolidate various prompt presets of a room into a set, which can be selected from either the general or private prompt catalog, enabling me to streamline the prompt selection process and customize the room experience more efficiently.
**Acceptance Criteria:**
1. Room creators can create and manage sets of prompt presets specific to their rooms.
2. These sets can be compiled from existing prompts in both the general and private prompt catalogs.
3. Within the room configuration, there is an option to select a prompt set from these catalogs, which can be applied to the room with ease.
4. The selected prompt set replaces the individual prompt selection process, simplifying the setup for the room creator.
6. Platform administrators have the flexibility to update or modify these sets as needed, with changes being reflected in real-time within the room.
7. The integration of prompt sets is designed to be user-friendly, with clear indications of which prompts are included in each set and easy navigation for selection and editing.
ARS-PLE ready for UX tests
2024-01-18
https://git.thm.de/arsnova/frag.jetzt/-/issues/1396
Enabling File and Picture Upload Options in Room Configuration for Chatbot In...
2023-12-21T13:54:55+01:00
Klaus-Dieter Quibeldey-Cirkel
Enabling File and Picture Upload Options in Room Configuration for Chatbot Interaction
**User Story:**
As a room creator, I want to have options in the room configuration stepper that allow participants to upload files and pictures when interacting with the chatbot, so that the conversations can be more dynamic, informati...
**User Story:**
As a room creator, I want to have options in the room configuration stepper that allow participants to upload files and pictures when interacting with the chatbot, so that the conversations can be more dynamic, informative, and engaging with the inclusion of multimedia content.
**Acceptance Criteria:**
1. The room configuration stepper includes a step to enable or disable file and picture uploads for chatbot interactions within the room.
2. When enabled, participants can securely upload files and pictures during their chatbot conversations, which the chatbot can acknowledge, respond to, or process as part of the interaction.
3. The upload feature supports a range of common file types (e.g., .jpg, .png for images; .pdf, .docx for documents) with size limits and security measures to prevent misuse.
5. All uploaded content is scanned for security risks (e.g., viruses, malware) to ensure the safety and integrity of the platform.
6. The interface for uploading files and pictures is user-friendly, with clear instructions and feedback on the upload process (e.g., progress bar, success/error messages).
ARS-PLE ready for UX tests
Ruben Bimberg
Ruben Bimberg
2024-01-18
https://git.thm.de/arsnova/frag.jetzt/-/issues/1395
Adaptation of the Onboarding Tour for ARS/PLE Modes
2023-11-23T08:07:12+01:00
Klaus-Dieter Quibeldey-Cirkel
Adaptation of the Onboarding Tour for ARS/PLE Modes
**User Story:**
As a new user of the platform, I want the onboarding tour to be tailored to the specific mode—ARS (Audience Response System) or PLE (Personalized Learning Environment)—that I choose, so that I can receive relevant guidan...
**User Story:**
As a new user of the platform, I want the onboarding tour to be tailored to the specific mode—ARS (Audience Response System) or PLE (Personalized Learning Environment)—that I choose, so that I can receive relevant guidance and information that enhances my understanding and initial experience with the chosen mode.
**Acceptance Criteria:**
1. The onboarding tour is responsive to the selection of either ARS or PLE mode, with the content and steps of the tour adjusting accordingly.
2. The tour for the ARS mode focuses on features, functionalities, and tips relevant to the Audience Response System, providing a deep understanding of how to interact with and benefit from this mode.
3. Conversely, the PLE mode tour emphasizes personalization features, learning tools, and resources unique to the Personalized Learning Environment.
4. The transition between the two modes can be made even during the onboarding tour, with the tour content dynamically changing to match the newly selected mode.
5. All instructions, tips, and guidance within the onboarding tour are clear, concise, and tailored to deliver the most relevant information for the selected mode.
6. The tour is designed to be engaging and interactive, ensuring new users are well-informed and comfortable with navigating the platform.