Savi

savi

Blockchain technology has long promised the creation of an independent financial ecosystem—one that operates free from the control of traditional finance institutions that have continually benefited from their extortion. Cryptocurrencies like Bitcoin and Ethereum emerged as decentralized alternatives to fiat, with the goal of democratizing finance and empowering individual sovereignty. Central to this mission is autonomy, transparency, and resistance to censorship. Stablecoins have become a natural extension to cryptocurrency. Acting as a proxy to real world assets, stablecoins hypothecate fiat currencies like the U.S. dollar on a blockchain. They have quickly emerged as a useful tool in the crypto ecosystem, offering a solution to volatility while still gaining the benefits of a 24/7 always-on system. Stablecoins have found a predominant product-market fit within crypto, becoming the backbone for payments, lending, and remittances. However, it is important to raise concerns about our reliance on stablecoins. Stablecoins have the ability to undermine blockchain autonomy and expose what should be a pristine network, isolated from traditional finance, to unduly force off-chain. Stablecoins as Settlement Layers Stablecoins like USDC are commonly perceived as cryptocurrencies, but in reality, they function closer to digital traditional payment gateways like Visa or Mastercard just on a blockchain. These stablecoins are backed by reserves held in traditional banks while the blockchain functions solely as an interchange protocol, similar to existing payment networks. This process closely resembles how traditional finance operates: funds are moved between accounts, with the underlying value anchored outside of the transaction. This setup is wildly different from the truly decentralized vision by early crypto pioneers. The canary in the coal mine for the chokehold stablecoins have on crypto is the comparison of transaction volume between stablecoins and native cryptocurrencies. Stablecoin transaction volume now even dominates the volume of native crypto assets like Bitcoin or Ether. The SVB Collapse and USDC The collapse of Silicon Valley Bank (SVB) in early 2023 is a stark reminder of how dangerous stablecoins deeply intertwined with traditional finance can be. As news broke of SVB’s potential insolvency, panic quickly spread through the crypto markets and fear around USDC’s peg arose. SVB’s collapse triggered a wave of USDC redemptions, essentially creating a bank run on USDC while USDC itself was experiencing a bank run on SVB. This sudden surge led to the de-pegging of USDC from $1 down to $0.84 at the lows. While the situation eventually stabilized, it exposed a critical vulnerability: stability on chain is directly tied to the health of our traditional banking system, something crypto aimed to wholly do away with. In fact, over a year and a half later, USDC has yet to fully recover in dominance against USDT. Stablecoin Dependencies Stablecoins destroy the independence and decentralization of blockchains through their reliance on traditional banking. This dependency extends beyond the mere financial aspect since stablecoins are vulnerable to regulatory control and influence, potentially comprising the censorship-resistant nature of crypto. As we know it today, U.S Treasury Secretary, Janet Yellen, could theoretically exert influence over the Ethereum network via the nearly $100B of stablecoins that exist on Ethereum. Moreover, stablecoin issuers possess the power to censor or block transactions, introducing risks to users that are not well advertised or communicated. This centralized control over such a critical and emerging sector of the crypto economy represents a significant deviation from a decentralized future without central banking. The U.S Government’s Stance on Stablecoins While many in the crypto industry are cheering on the United States government’s new found love of stablecoins, maybe we should raise our eyebrows slightly. Given the government’s previously critical stance on cryptocurrencies, this apparent contradiction lends credence to the idea that stablecoins can be a Trojan Horse for censorship and CBDCs. It is also understandable why the U.S sees stablecoins as an important sector. Stablecoins are a useful vehicle for exporting dollar dominance into the digital age. In fact, stablecoins are the 16th largest holder of US Treasuries. By encouraging the adoption of dollar-pegged stablecoins, the government can extend its monopoly on the monetary system it has held tightly since the Bretton Woods agreement in 1944. Economic Implications and Blockchain Security Every dollar invested in stablecoins on a blockchain network represents a missed opportunity for investment in the network's native asset. Diversion of capital decreases the economic security of the underlying blockchain, particularly in Proof of Stake systems. As the market cap of stablecoins continues to grow, that is value that in theory could have been attributed to the economic security of Ethereum itself. Furthermore, The fees generated from stablecoin transactions often flow to centralized entities rather than the network operators themselves. This outflow of value undermines the long term revenue for network participants and the fees paid by users. A simple example is an ETH payment on Ethereum costing $1 in fees while a USDT payment on Ethereum may cost $30 at the same gas price. The situation at hand not only drives users towards centralized stablecoins but to centralized networks as well. It is a bleak outlook if the future of crypto is purely centralized stablecoins on centralized L2s operated by centralized exchanges. Potential Solutions Fortunately, there are new economic primitives emerging that mitigate these aforementioned centralized forces imposed by stablecoins. Alternatives include ETH derivatives such as Reflexer’s RAI token, commodity-backed assets, and native crypto assets, each offering unique approaches to achieving stability without relying on TradFi. It’s also worth noting that given most stablecoins do not distribute treasury yield or track CPI, they are not truly “stable” in the strictest sense and instead lose purchasing power. This has opened the door for “flatcoins” like SPOT or NUON to further innovate although at similar dependencies for centralization. Novel approaches to achieving stability without traditional backing are constantly being developed. These include liquidity-incentivized pools, dynamic supply adjustment mechanisms, and cross-chain collateralization strategies. While none have yet achieved the dominance of fiat-backed stablecoins, they represent important steps towards a more autonomous blockchain ecosystem. The stablecoin landscape presents a variety of approaches, each with its own set of trade-offs. Treasury-backed stablecoins offer high stability but are highly centralized, while synthetic stablecoins introduce potential points of failure through reliance on exchanges or custodians. Decentralized alternatives aim for greater autonomy but have faced challenges with capital efficiency. In this context, the concept of an energy-based stablecoin alternative emerges as a promising innovation. By tying value to computational work and energy costs, it offers a unique solution that addresses many of the shortcomings of existing designs. This approach provides a tangible, real-world basis for pricing, potentially enhancing stability without sacrificing decentralization. It also aligns well with the increasingly digital and energy-conscious global economy. Conclusion Stablecoins are set to be an enormous product market fit for blockchain technology and will accelerate the use of crypto adoption globally. However, it is crucial to recognize how these assets jeopardize some of the most desirable and unique properties of cryptocurrency, namely decentralization, censorship-resistance, and economic independence. This fundamental conflict highlights the friction of how people are realistically using blockchain technology today and how crypto idealists see the future. As the crypto industry continues to evolve, it is critical that we evaluate all dependencies on centralized systems and work towards solutions that give financial sovereignty back to users. Blockchain technology has long promised the creation of an independent financial ecosystem—one that operates free from the control of traditional finance institutions that have continually benefited from their extortion. Cryptocurrencies like Bitcoin and Ethereum emerged as decentralized alternatives to fiat, with the goal of democratizing finance and empowering individual sovereignty. Central to this mission is autonomy, transparency, and resistance to censorship. Stablecoins have become a natural extension to cryptocurrency. Acting as a proxy to real world assets, stablecoins hypothecate fiat currencies like the U.S. dollar on a blockchain. They have quickly emerged as a useful tool in the crypto ecosystem, offering a solution to volatility while still gaining the benefits of a 24/7 always-on system. Stablecoins have found a predominant product-market fit within crypto, becoming the backbone for payments, lending, and remittances. However, it is important to raise concerns about our reliance on stablecoins. Stablecoins have the ability to undermine blockchain autonomy and expose what should be a pristine network, isolated from traditional finance, to unduly force off-chain. Stablecoins as Settlement Layers Stablecoins like USDC are commonly perceived as cryptocurrencies, but in reality, they function closer to digital traditional payment gateways like Visa or Mastercard just on a blockchain. These stablecoins are backed by reserves held in traditional banks while the blockchain functions solely as an interchange protocol, similar to existing payment networks. This process closely resembles how traditional finance operates: funds are moved between accounts, with the underlying value anchored outside of the transaction. This setup is wildly different from the truly decentralized vision by early crypto pioneers. The canary in the coal mine for the chokehold stablecoins have on crypto is the comparison of transaction volume between stablecoins and native cryptocurrencies. Stablecoin transaction volume now even dominates the volume of native crypto assets like Bitcoin or Ether. The SVB Collapse and USDC The collapse of Silicon Valley Bank (SVB) in early 2023 is a stark reminder of how dangerous stablecoins deeply intertwined with traditional finance can be. As news broke of SVB’s potential insolvency, panic quickly spread through the crypto markets and fear around USDC’s peg arose. SVB’s collapse triggered a wave of USDC redemptions, essentially creating a bank run on USDC while USDC itself was experiencing a bank run on SVB. This sudden surge led to the de-pegging of USDC from $1 down to $0.84 at the lows. While the situation eventually stabilized, it exposed a critical vulnerability: stability on chain is directly tied to the health of our traditional banking system, something crypto aimed to wholly do away with. In fact, over a year and a half later, USDC has yet to fully recover in dominance against USDT. Stablecoin Dependencies Stablecoins destroy the independence and decentralization of blockchains through their reliance on traditional banking. This dependency extends beyond the mere financial aspect since stablecoins are vulnerable to regulatory control and influence, potentially comprising the censorship-resistant nature of crypto. As we know it today, U.S Treasury Secretary, Janet Yellen, could theoretically exert influence over the Ethereum network via the nearly $100B of stablecoins that exist on Ethereum. Moreover, stablecoin issuers possess the power to censor or block transactions, introducing risks to users that are not well advertised or communicated. This centralized control over such a critical and emerging sector of the crypto economy represents a significant deviation from a decentralized future without central banking. The U.S Government’s Stance on Stablecoins The U.S Government’s Stance on Stablecoins While many in the crypto industry are cheering on the United States government’s new found love of stablecoins, maybe we should raise our eyebrows slightly. Given the government’s previously critical stance on cryptocurrencies, this apparent contradiction lends credence to the idea that stablecoins can be a Trojan Horse for censorship and CBDCs. It is also understandable why the U.S sees stablecoins as an important sector. Stablecoins are a useful vehicle for exporting dollar dominance into the digital age. In fact, stablecoins are the 16th largest holder of US Treasuries. By encouraging the adoption of dollar-pegged stablecoins, the government can extend its monopoly on the monetary system it has held tightly since the Bretton Woods agreement in 1944. Economic Implications and Blockchain Security Every dollar invested in stablecoins on a blockchain network represents a missed opportunity for investment in the network's native asset. Diversion of capital decreases the economic security of the underlying blockchain, particularly in Proof of Stake systems. As the market cap of stablecoins continues to grow, that is value that in theory could have been attributed to the economic security of Ethereum itself. Furthermore, The fees generated from stablecoin transactions often flow to centralized entities rather than the network operators themselves. This outflow of value undermines the long term revenue for network participants and the fees paid by users. A simple example is an ETH payment on Ethereum costing $1 in fees while a USDT payment on Ethereum may cost $30 at the same gas price. The situation at hand not only drives users towards centralized stablecoins but to centralized networks as well. It is a bleak outlook if the future of crypto is purely centralized stablecoins on centralized L2s operated by centralized exchanges. Potential Solutions Fortunately, there are new economic primitives emerging that mitigate these aforementioned centralized forces imposed by stablecoins. Alternatives include ETH derivatives such as Reflexer’s RAI token, commodity-backed assets, and native crypto assets, each offering unique approaches to achieving stability without relying on TradFi. It’s also worth noting that given most stablecoins do not distribute treasury yield or track CPI, they are not truly “stable” in the strictest sense and instead lose purchasing power. This has opened the door for “flatcoins” like SPOT or NUON to further innovate although at similar dependencies for centralization. Novel approaches to achieving stability without traditional backing are constantly being developed. These include liquidity-incentivized pools, dynamic supply adjustment mechanisms, and cross-chain collateralization strategies. While none have yet achieved the dominance of fiat-backed stablecoins, they represent important steps towards a more autonomous blockchain ecosystem. The stablecoin landscape presents a variety of approaches, each with its own set of trade-offs. Treasury-backed stablecoins offer high stability but are highly centralized, while synthetic stablecoins introduce potential points of failure through reliance on exchanges or custodians. Decentralized alternatives aim for greater autonomy but have faced challenges with capital efficiency. In this context, the concept of an energy-based stablecoin alternative emerges as a promising innovation. By tying value to computational work and energy costs, it offers a unique solution that addresses many of the shortcomings of existing designs. This approach provides a tangible, real-world basis for pricing, potentially enhancing stability without sacrificing decentralization. It also aligns well with the increasingly digital and energy-conscious global economy. Conclusion Stablecoins are set to be an enormous product market fit for blockchain technology and will accelerate the use of crypto adoption globally. However, it is crucial to recognize how these assets jeopardize some of the most desirable and unique properties of cryptocurrency, namely decentralization, censorship-resistance, and economic independence. This fundamental conflict highlights the friction of how people are realistically using blockchain technology today and how crypto idealists see the future. As the crypto industry continues to evolve, it is critical that we evaluate all dependencies on centralized systems and work towards solutions that give financial sovereignty back to users.

The Paradox of Stablecoins: The Savior or Achilles’ Heel of Crypto?

VAUL3 proudly unveils its innovative peer-to-peer (P2P) smart saving ecosystem, a platform designed to bridge the gap between traditional finance and the rapidly evolving decentralized finance (DeFi) landscape. As financial markets continue to adapt to the digital age, VAUL3 is poised to lead this transformation with a mission to democratize financial growth, empowering users to create, manage, and invest in yield strategies with unprecedented ease and security. The Future of Finance: VAUL3 ’s Vision and Innovation At the core of VAUL3 is a user-centric platform that leverages blockchain technology to ensure transparency, security, and efficiency in all transactions. With traditional financial systems often mired in complexity and exclusivity, VAUL3 seeks to break down these barriers, offering tools that make sophisticated financial strategies accessible to all. The platform's innovative approach is anchored in simplicity and user empowerment. By providing no-code tools, VAUL3 allows both creators and investors to engage with the platform regardless of their technical expertise. Whether you're an experienced DeFi strategist or a newcomer to the space, VAUL3 equips you with everything needed to succeed in a decentralized financial ecosystem. For creators, VAUL3 offers streamlined fundraising processes, automated fund management through smart contracts, and advanced analytics to refine strategies and optimize performance. Investors, on the other hand, benefit from diverse investment opportunities tailored to various risk profiles, comprehensive risk mitigation features, and real-time performance tracking to monitor and manage their portfolios effectively. Launching the $VAUL3 Token: A Strategic Investment Opportunity As part of its platform launch, VAUL3 is excited to announce the presale of its native utility token, $VAUL3. This token is designed to be the engine that drives the VAUL3 ecosystem, facilitating transactions, incentivizing participation, and enabling access to premium features and governance roles within the platform. The $VAUL3 token presale represents a significant opportunity for early investors to participate in the platform’s growth from the outset. Priced at an advantageous rate, the first round of the presale is set at $0.0064 per token. Presale 1 Details Total Supply: 999,999,999 $VAUL3 tokens
Presale 1 Supply: 39,999,999 $VAUL3 tokens
Price: $0.0064 per token
Minimum Investment: $100
Maximum Investment: $49,000 This initial presale is a pivotal moment for those looking to be part of the next wave of financial innovation. The tokenomics of $VAUL3 have been meticulously designed to create long-term value for all participants. Subsequent presale rounds will see an increase in token price, reflecting the growing value and utility of the $VAUL3 token as the platform expands. A Vision for Long-Term Growth and Value Creation The launch of VAUL3 is not just about introducing a new platform; it’s about setting the stage for the future of decentralized finance. The VAUL3 ecosystem is built with scalability and sustainability in mind, ensuring that it can grow alongside its user base and continue to offer value well into the future. “VAUL3 is set to redefine how individuals and institutions engage with yield strategies, providing a platform that is as secure as it is accessible,” said CEO & Co-founder of VAUL3. “Our presale is a unique opportunity for investors to get in on the ground floor of a platform poised to make a significant impact in the DeFi space. We are excited to see how our community will grow and evolve as we continue to innovate and push the boundaries of what’s possible in finance.” Join the Revolution in Decentralized Finance The VAUL3 team invites investors, creators, and the broader financial community to join in this groundbreaking initiative. With the launch of the $VAUL3 token presale, there has never been a better time to become part of a platform that is set to redefine the future of finance. Secure your $VAUL3 tokens today and be part of the future of finance. For more information on the presale and to join our growing community, visit www.vaul3.com/vaul3-token This story was distributed as a release by ZEX MEDIA under HackerNoon’s Brand As An Author Program. Learn more about the program here. VAUL3 proudly unveils its innovative peer-to-peer (P2P) smart saving ecosystem , a platform designed to bridge the gap between traditional finance and the rapidly evolving decentralized finance (DeFi) landscape. As financial markets continue to adapt to the digital age, VAUL3 is poised to lead this transformation with a mission to democratize financial growth, empowering users to create, manage, and invest in yield strategies with unprecedented ease and security. VAUL3 VAUL3 peer-to-peer (P2P) smart saving ecosystem peer-to-peer (P2P) smart saving ecosystem The Future of Finance: VAUL3 ’s Vision and Innovation At the core of VAUL3 is a user-centric platform that leverages blockchain technology to ensure transparency, security, and efficiency in all transactions. With traditional financial systems often mired in complexity and exclusivity, VAUL3 seeks to break down these barriers, offering tools that make sophisticated financial strategies accessible to all. The platform's innovative approach is anchored in simplicity and user empowerment. By providing no-code tools, VAUL3 allows both creators and investors to engage with the platform regardless of their technical expertise. Whether you're an experienced DeFi strategist or a newcomer to the space, VAUL3 equips you with everything needed to succeed in a decentralized financial ecosystem. For creators, VAUL3 offers streamlined fundraising processes, automated fund management through smart contracts, and advanced analytics to refine strategies and optimize performance. Investors, on the other hand, benefit from diverse investment opportunities tailored to various risk profiles, comprehensive risk mitigation features, and real-time performance tracking to monitor and manage their portfolios effectively. Launching the $VAUL3 Token: A Strategic Investment Opportunity As part of its platform launch, VAUL3 is excited to announce the presale of its native utility token, $VAUL3 . This token is designed to be the engine that drives the VAUL3 ecosystem, facilitating transactions, incentivizing participation, and enabling access to premium features and governance roles within the platform. $VAUL3 $VAUL3 The $VAUL3 token presale represents a significant opportunity for early investors to participate in the platform’s growth from the outset. Priced at an advantageous rate, the first round of the presale is set at $0.0064 per token. Presale 1 Details Total Supply: 999,999,999 $VAUL3 tokens Presale 1 Supply: 39,999,999 $VAUL3 tokens Price: $0.0064 per token Minimum Investment: $100 Maximum Investment: $49,000 Total Supply: 999,999,999 $VAUL3 tokens Presale 1 Supply: 39,999,999 $VAUL3 tokens Price: $0.0064 per token Minimum Investment: $100 Maximum Investment: $49,000 This initial presale is a pivotal moment for those looking to be part of the next wave of financial innovation. The tokenomics of $VAUL3 have been meticulously designed to create long-term value for all participants. Subsequent presale rounds will see an increase in token price, reflecting the growing value and utility of the $VAUL3 token as the platform expands. A Vision for Long-Term Growth and Value Creation The launch of VAUL3 is not just about introducing a new platform; it’s about setting the stage for the future of decentralized finance. The VAUL3 ecosystem is built with scalability and sustainability in mind, ensuring that it can grow alongside its user base and continue to offer value well into the future. “VAUL3 is set to redefine how individuals and institutions engage with yield strategies, providing a platform that is as secure as it is accessible,” said CEO & Co-founder of VAUL3. “Our presale is a unique opportunity for investors to get in on the ground floor of a platform poised to make a significant impact in the DeFi space. We are excited to see how our community will grow and evolve as we continue to innovate and push the boundaries of what’s possible in finance.” Join the Revolution in Decentralized Finance The VAUL3 team invites investors, creators, and the broader financial community to join in this groundbreaking initiative. With the launch of the $VAUL3 token presale, there has never been a better time to become part of a platform that is set to redefine the future of finance. Secure your $VAUL3 tokens today and be part of the future of finance. For more information on the presale and to join our growing community, visit www.vaul3.com/vaul3-token www.vaul3.com/vaul3-token www.vaul3.com/vaul3-token This story was distributed as a release by ZEX MEDIA under HackerNoon’s Brand As An Author Program. Learn more about the program here. This story was distributed as a release by ZEX MEDIA under HackerNoon’s Brand As An Author Program. Learn more about the program here. This story was distributed as a release by ZEX MEDIA under HackerNoon’s Brand As An Author Program. Learn more about the program here. This story was distributed as a release by ZEX MEDIA under HackerNoon’s Brand As An Author Program. Learn more about the program here . here here

VAUL3 Announces the Launch of Its P2P Smart Saving Ecosystem and $VAUL3 Token Presale

VAUL3 তার P2P স্মার্ট সেভিং ইকোসিস্টেম এবং $VAUL3 টোকেন প্রিসেল চালু করার ঘোষণা দিয়েছে

VAUL3 gibt den Start seines P2P-Smart-Saving-Ökosystems und den Vorverkauf des $VAUL3-Tokens bekannt

VAUL3 anuncia el lanzamiento de su ecosistema de ahorro inteligente P2P y la preventa del token $VAUL3

VAUL3 annonce le lancement de son écosystème d'épargne intelligent P2P et la prévente du jeton $VAUL3

VAUL3 ने अपने P2P स्मार्ट सेविंग इकोसिस्टम और $VAUL3 टोकन प्रीसेल के लॉन्च की घोषणा की

VAUL3 が P2P スマート貯蓄エコシステムと $VAUL3 トークンの先行販売の開始を発表

VAUL3, P2P 스마트 저축 생태계 출시 및 $VAUL3 토큰 사전 판매 발표

VAUL3 anuncia o lançamento de seu ecossistema de economia inteligente P2P e pré-venda do token $VAUL3

VAUL3 объявляет о запуске своей экосистемы интеллектуальных сбережений P2P и предварительной продаже токенов $VAUL3

VAUL3, P2P Akıllı Tasarruf Ekosisteminin ve $VAUL3 Token Ön Satışının Başlatıldığını Duyurdu

VAUL3 công bố ra mắt hệ sinh thái tiết kiệm thông minh P2P và đợt bán trước token $VAUL3

VAUL3 宣布推出 P2P 智能储蓄生态系统并启动 $VAUL3 代币预售

Frequent alerts can lead to notification fatigue, causing users to disengage. I knew that by batching notifications, we could maintain user attention and promote sustained interaction with our platform. However, setting up an efficient batching system required some complex workflows and significant development resources. You can explore the hands-on implementation through the following links:

GitHub: Github Social App
Deployed Application: Github Project Demo The Batching Concept The concept of batching notifications involves several technical components: Aggregation Engine: This engine efficiently aggregates related notifications, such as likes, shares, and comments, based on metadata attributes. For example, Instagram separates notifications for story likes and comments to ensure clarity.
Batching Window: The batching window can be fixed (e.g., every 30 minutes) or dynamic (e.g., user-specific intervals). For instance, LinkedIn batches email alerts for new messages every 30 minutes, while Google Docs batches comments based on user activity.
Scheduling: This determines the optimal time for delivering notifications. It could be immediately, at the end of a batching window, or at a strategic time. Many SaaS companies send a daily digest of activities the following morning.
Batched Message Presentation: Options range from simple counters to detailed summaries, balancing informativeness and engagement without overwhelming the user. For example, "Patrick and 3 others liked your photo" versus listing all activities.
Cancelling Aggregation: This adjusts the aggregation counter for counter-activities within the batching window. For example, if a user likes and then unlikes a post within the batching window, the counter is adjusted accordingly. Building the Demo Application To demonstrate this, I chose to build a React application using SuprSend's robust notification infrastructure. Here’s a step-by-step account of how I did it: Setting Up the Project:


I started by creating a React application and integrating SuprSend's SDK.


I set up a SuprSend account and ensured successful event calls were made to SuprSend with the necessary details.




Identifying Triggers:


I identified recurring events suitable for batching. For this demo, I used the Like_Event whenever someone likes a post. Using SuprSend’s JavaScript SDK, I sent these events to the backend.








Configuring Batch Parameters:

Using SuprSend’s workflow builder, I configured the batch parameters:


Batch Window: I set a fixed batch window of 15 seconds for quick testing.


Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately.


Retain Batch Events: I chose to retain 15 objects in the array for this demo.






Creating Notification Templates:


I created templates in SuprSend’s editor using Handlebar Helpers to format notifications based on the batched event count. The template looked like this:
{{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} The final result? A neat batched notification. Implementing Throttling While batching reduces the total number of notifications, I realized that introducing throttling could further enhance user experience by limiting the frequency of notifications. Users would not be overwhelmed even with multiple batches by setting an upper limit on daily notifications. Will tell in next tutorial how I implemented that. The Result The final demo application successfully demonstrated how batching notifications could improve user engagement. You can explore the hands-on implementation through the following links: GitHub: Github Social App
Deployed Application: Github Project Demo Frequent alerts can lead to notification fatigue, causing users to disengage. I knew that by batching notifications, we could maintain user attention and promote sustained interaction with our platform. However, setting up an efficient batching system required some complex workflows and significant development resources. You can explore the hands-on implementation through the following links: GitHub: Github Social App
Deployed Application: Github Project Demo You can explore the hands-on implementation through the following links: GitHub: Github Social App Deployed Application: Github Project Demo GitHub: Github Social App Social App Deployed Application: Github Project Demo Github Project Demo The Batching Concept The concept of batching notifications involves several technical components: Aggregation Engine: This engine efficiently aggregates related notifications, such as likes, shares, and comments, based on metadata attributes. For example, Instagram separates notifications for story likes and comments to ensure clarity. Batching Window: The batching window can be fixed (e.g., every 30 minutes) or dynamic (e.g., user-specific intervals). For instance, LinkedIn batches email alerts for new messages every 30 minutes, while Google Docs batches comments based on user activity. Scheduling: This determines the optimal time for delivering notifications. It could be immediately, at the end of a batching window, or at a strategic time. Many SaaS companies send a daily digest of activities the following morning. Batched Message Presentation: Options range from simple counters to detailed summaries, balancing informativeness and engagement without overwhelming the user. For example, "Patrick and 3 others liked your photo" versus listing all activities. Cancelling Aggregation: This adjusts the aggregation counter for counter-activities within the batching window. For example, if a user likes and then unlikes a post within the batching window, the counter is adjusted accordingly. Aggregation Engine : This engine efficiently aggregates related notifications, such as likes, shares, and comments, based on metadata attributes. For example, Instagram separates notifications for story likes and comments to ensure clarity. Aggregation Engine Batching Window : The batching window can be fixed (e.g., every 30 minutes) or dynamic (e.g., user-specific intervals). For instance, LinkedIn batches email alerts for new messages every 30 minutes, while Google Docs batches comments based on user activity. Batching Window Scheduling : This determines the optimal time for delivering notifications. It could be immediately, at the end of a batching window, or at a strategic time. Many SaaS companies send a daily digest of activities the following morning. Scheduling Batched Message Presentation : Options range from simple counters to detailed summaries, balancing informativeness and engagement without overwhelming the user. For example, "Patrick and 3 others liked your photo" versus listing all activities. Batched Message Presentation Cancelling Aggregation : This adjusts the aggregation counter for counter-activities within the batching window. For example, if a user likes and then unlikes a post within the batching window, the counter is adjusted accordingly. Cancelling Aggregation Building the Demo Application To demonstrate this, I chose to build a React application using SuprSend's robust notification infrastructure. Here’s a step-by-step account of how I did it: Setting Up the Project:


I started by creating a React application and integrating SuprSend's SDK.


I set up a SuprSend account and ensured successful event calls were made to SuprSend with the necessary details. Identifying Triggers:


I identified recurring events suitable for batching. For this demo, I used the Like_Event whenever someone likes a post. Using SuprSend’s JavaScript SDK, I sent these events to the backend. Configuring Batch Parameters:

Using SuprSend’s workflow builder, I configured the batch parameters:


Batch Window: I set a fixed batch window of 15 seconds for quick testing.


Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately.


Retain Batch Events: I chose to retain 15 objects in the array for this demo. Creating Notification Templates:


I created templates in SuprSend’s editor using Handlebar Helpers to format notifications based on the batched event count. The template looked like this:
{{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} Setting Up the Project: I started by creating a React application and integrating SuprSend's SDK.


I set up a SuprSend account and ensured successful event calls were made to SuprSend with the necessary details. Setting Up the Project : Setting Up the Project I started by creating a React application and integrating SuprSend's SDK. I set up a SuprSend account and ensured successful event calls were made to SuprSend with the necessary details. I started by creating a React application and integrating SuprSend's SDK. I started by creating a React application and integrating SuprSend's SDK. I set up a SuprSend account and ensured successful event calls were made to SuprSend with the necessary details. I set up a SuprSend account and ensured successful event calls were made to SuprSend with the necessary details. Identifying Triggers: I identified recurring events suitable for batching. For this demo, I used the Like_Event whenever someone likes a post. Using SuprSend’s JavaScript SDK, I sent these events to the backend. Identifying Triggers : Identifying Triggers I identified recurring events suitable for batching. For this demo, I used the Like_Event whenever someone likes a post. Using SuprSend’s JavaScript SDK, I sent these events to the backend. I identified recurring events suitable for batching. For this demo, I used the Like_Event whenever someone likes a post. Using SuprSend’s JavaScript SDK, I sent these events to the backend. I identified recurring events suitable for batching. For this demo, I used the Like_Event whenever someone likes a post. Using SuprSend’s JavaScript SDK, I sent these events to the backend. Like_Event Configuring Batch Parameters: Using SuprSend’s workflow builder, I configured the batch parameters:


Batch Window: I set a fixed batch window of 15 seconds for quick testing.


Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately.


Retain Batch Events: I chose to retain 15 objects in the array for this demo. Configuring Batch Parameters : Configuring Batch Parameters Using SuprSend’s workflow builder, I configured the batch parameters:


Batch Window: I set a fixed batch window of 15 seconds for quick testing.


Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately.


Retain Batch Events: I chose to retain 15 objects in the array for this demo. Using SuprSend’s workflow builder, I configured the batch parameters: Batch Window: I set a fixed batch window of 15 seconds for quick testing.


Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately.


Retain Batch Events: I chose to retain 15 objects in the array for this demo. Batch Window: I set a fixed batch window of 15 seconds for quick testing. Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately. Retain Batch Events: I chose to retain 15 objects in the array for this demo. Batch Window: I set a fixed batch window of 15 seconds for quick testing. Batch Window : I set a fixed batch window of 15 seconds for quick testing. Batch Window Batch Key: I defined a custom batch key based on the userName variable to batch likes on different posts separately. Batch Key : I defined a custom batch key based on the userName variable to batch likes on different posts separately. Batch Key Retain Batch Events: I chose to retain 15 objects in the array for this demo. Retain Batch Events : I chose to retain 15 objects in the array for this demo. Retain Batch Events Creating Notification Templates: I created templates in SuprSend’s editor using Handlebar Helpers to format notifications based on the batched event count. The template looked like this:
{{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} Creating Notification Templates : Creating Notification Templates I created templates in SuprSend’s editor using Handlebar Helpers to format notifications based on the batched event count. The template looked like this:
{{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} I created templates in SuprSend’s editor using Handlebar Helpers to format notifications based on the batched event count. The template looked like this: {{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} I created templates in SuprSend’s editor using Handlebar Helpers to format notifications based on the batched event count. The template looked like this: {{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} {{#compare $batched_events_count '>' 1}} 
{{ $batched_events.[0].username }} and {{ subtract $batched_events_count 1 }} others liked your post. 
{{else}} 
{{ $batched_events.[0].username }} liked your post.
{{/compare}} The final result? The final result? The final result? A neat batched notification. Implementing Throttling While batching reduces the total number of notifications, I realized that introducing throttling could further enhance user experience by limiting the frequency of notifications. Users would not be overwhelmed even with multiple batches by setting an upper limit on daily notifications. Will tell in next tutorial how I implemented that. The Result The final demo application successfully demonstrated how batching notifications could improve user engagement. You can explore the hands-on implementation through the following links: GitHub: Github Social App Deployed Application: Github Project Demo GitHub: Github Social App Github Social App Deployed Application: Github Project Demo Github Project Demo

Creating Batched Notifications in a Dedicated Time Window for Better UX and Saving Notification Cost

Welcome to Daylight Savings Recovery Week – a period of days when most of the world expresses shock, dismay, despair, and the other 5 stages of grief about waking up an hour early and generally being cranky. Like so many things in 2024, the act of saving daylight (whatever that might actually mean) has become increasingly polarizing (and somewhat partisan) with both citizens and politicians clamoring for an end to the madness of changing our clocks twice a year. on a bill signed in 2022 that would keep daylight savings time permanently in place. Colorado is awaiting federal approval would permanently end daylight savings time. A new bill in California got a bill passed in the Senate that would put an end to DST, but it wasn’t taken up in the house. In the US Congress, a senator from Oklahoma , another Republican – from Utah this time – reintroduced a bill to make DST permanent, or at least allow states to do so if they chose. Meanwhile, over in the House In all, there are that relate to the abolishment of, or standardization of, daylight saving time. 500 bills (between state and federal legislatures) currently in play I’m sharing this by way of making two points: Lots of people have: an opinion on daylight saving time the authority to try to change it the desire to change it now There is no consistency, no unified plan on how to do this. The result of this can be described by words ranging from “absurd” to “disastrous.” “Absurd” because it’s entirely possible that, in the current situation, one state would abolish DST, while another would permanently stay on it. Meaning you could travel from Ohio at 6:00 am (with a bill currently under consideration to remain on DST), drive 1 hour to Pennsylvania where it would be 8:00 am (because they’ve continued to change their clocks seasonally); and arrive 4 hours later in New York, where the time is… no, not 12:00 pm, but instead, it would be 11:00 am. I’m probably exaggerating a little because, as I understand it, states can pass all the DST bills they want but it will require a federal ruling to allow that law to take effect. Or something like that. I’m neither a lawyer nor a politician so I’m not 100% sure. Let’s face it, it’s 2024. The push to remote work that started with the pandemic has caused many of us to feel that time is mostly meaningless. Things get done, people are awake (or asleep) whenever it suits them, and the number on the clock has less bearing on our lives than it once did. Nevertheless, the current rash of polarized opinions related to changing daylight savings time could make life (or at least certain times of day) very hard for folks who observe solar-centric, rather than clock-centric, times. , which helps developers easily incorporate and display so-called “Jewish” times (zmanim) into their websites. But here’s a quick primer for those who haven’t followed along: I’ve been writing about my work using PHP Zmanim It’s difficult to understate the importance of accurate time calculations on the Jewish religious day, week, and life. It figures into everything from when a fast starts and ends; to when it’s too late (or too early) to say certain prayers; to the moment when simply owning (let alone eating) bread is forbidden. “What time are afternoon prayers?” To understand that, we have to first understand what defines “day” (versus “night”); how to understand what demarcation separates “morning” from “afternoon”; how to effectively divide the day into parts; and so much more. Here’s a sampling of sun-specific times: Before sunrise, (the moment when the top edge of the disk of the sun touches the horizon) comes “dawn”, the moment when light can first be seen shining over the horizon. No prayers can happen before dawn. At a specific time between dawn and sunrise, it’s permissible to put on tallit and tefillin. a “seasonal hour” is determined by taking the amount of daylight (from sunrise to sunset) and dividing it into 12 equal parts. Afternoon prayers cannot be said until 6.5 seasonal hours after sunrise. My point is not to bog you down with complex (and, to the non-religious reader, seemingly whimsical) calculations. My point is that there is a significant population of folks (a rough estimate is 3.1 million people) for whom this DOES matter. And that doesn’t even begin to consider that the 5 salah (Muslim prayers) times are equally dependent on the sun. How does the current kerfluffle about DST factor in? let’s consider (all times are given for my location in Cleveland Ohio. Your Sunlight May Vary!): On Nov 5, the, time “falls” back – so that what WOULD have been an 8 am sunrise becomes 7 am. But sunrise, and sunlight, continue moving forward. In fact, the day gets progressively later all the way to Dec 31, when it hits 7:53 am Sunrise sits at that point for another 8 days, until it starts rolling back (getting earlier) starting January 8. But consider that without that shift in time, sunrise on November 6 would proceed forward – eventually stopping just shy of 9 am. For the observant Jewish person, that means dawn comes at almost 7:30 am. The earliest time for tallit and tefillin would be 7:56 am. A typical morning service runs between 30-45 minutes, meaning that most folks won’t get out of services until 8:45 am. For a person who has an early start to their day, or has to travel to the office; or get kids to school, this makes attending prayers almost impossible. On the other hand, what happens if we don’t ever go back to daylight saving? On March 9 the clocks “spring forward” so that the 6:47 am sunrise becomes 7:45 am on March 10. But sunrise continues to get earlier, all the way until June 8, when it hits 5:52 am. And sunrise sits at that time for 12 days, all the way until it finally starts to move forward again on June 20. But without the DST adjustment, sunrise would come at 4:52 am. You might expect, at this point, that I’d do a big reveal and show that sunSET is impacted. That’s not the case. Back in November, sunset goes from 6:18 pm on November 4 to 5:17 pm the following day. It keeps getting earlier until December 5 when it hits 4:56 pm. Now, I’ll freely admit that I think sunset at 5 pm kind of sucks. But I’m willing to accept it far more than having to manage prayers at almost 9 am. Meanwhile, on the other side of the equation, the latest sunset hits on June 20, at 9:04 pm. If you are completely turned around at that point, join the club. I had to do a few exports from and some manual sorting to get it all straight: timeanddate.com If we never sprang forward into daylight saving time: The earliest sunrise would be 4:49 am (on Jun 13) The latest sunrise would be 7:51 am (Jan 9) The earliest sunset would be 4:56 pm (Dec 4) The latest sunset would be 8:04 pm (Jul 4) Meanwhile, if we never fell back – meaning permanent daylight saving time: The earliest sunrise would be 5:49 am (Jun 13) The latest sunrise would be 8:51 am (Jan 9) The earliest sunset would be 5:56 pm (Dec 4) The latest sunset would be 9:04 pm (Jul 4) It’s worth clarifying what DST is giving us now: The earliest sunrise is 5:49 am (Jun 13) The latest sunrise is 7:59 am (Nov 5) The earliest sunset is 4:56 pm (Dec 4) The latest sunset is 9:04 pm (Jul 4) With all of that said, my opinion is that the DST system doesn’t offer very much benefit for the trouble. Moreover, as an orthodox Jew – for whom prayers (and therefore the cycles of sunrise and sunset) matter more than the time on a clock – remaining on standard (non-DST) time would be my choice. On the one hand, I’m in luck. While, under Federal law, states can opt out of DST, they cannot choose to remain permanently on DST without approval. On the other, it seems this is the less popular choice. In recent years, to make DST permanent have been considered in various states across the country. 19 states have passed those bills, which would immediately take effect the moment the federal government approves it. over 650 bills But for the moment, things appear to be staying put. I guess I’ll try not to lose any sleep over it. Share this: Also published here

What Daylight Are We Saving? Time Data Series

My initial encounter with ChatGPT was driven by sheer curiosity when it was released. I was particularly drawn to articles discussing code generation, even though I don't personally write code. And I initially believed that feature wouldn't be relevant to me. However, there came a moment when I found myself working on one UML diagram that I wasn't too eager to create. In my quest to simplify or expedite the process, I turned to various methods. I usually craft UML diagrams using PlantUML code... here, the realization struck me: this is a form of code, and code can be generated. Thus, ChatGPT evolved into a valuable tool for my diagram creation, and now I'm excited to share my experiences with you. A Few Words About UML stands for , and it serves as a language for specifying, visualizing, constructing, and documenting artifacts of software systems. UML draws from concepts in object-oriented programming languages; having development knowledge is beneficial for UML modeling but not mandatory. UML Unified Modeling Language It is primarily designed to describe aspects of software design, but it also offers a set of behavioral diagrams to model business or user requirements. In my role, I engage with IT business processes, specifically focusing on modeling them for analysis, optimization, and translating them into a clear format for developers to implement. When it comes to modeling, I employ various approaches and tools, one of which is UML. Here's the beauty of it — you don't have to be a software engineer to utilize UML. I use it to navigate through business processes implemented with software, but it goes beyond that — it's remarkably versatile and can be applied by various roles at different stages of development. All in all, give it a try; you might just find yourself loving it. Draw vs Code Like any diagram, you can easily draw it — it's simple, convenient, quick, and provides a clear view of the process and the outcome. Numerous graphic tools are available for this purpose, and I do use them. However, more often than not, I choose the "code" approach, thanks to PlantUML. , an open-source modeling language and tool, allows you to generate diagrams based on plain text language. In simpler terms, you can write code in PlantUML to create diagrams. PlantUML I enjoy sketching diagrams in my 'creative mode,' but more often than not, I either convert them into code afterward or write the code directly. There are two main reasons for this: You can insert, move, and delete blocks in your diagrams much more efficiently Changing the order of blocks in drawn diagrams often leads to modifications across the entire diagram, affecting other blocks and links. When working with code, you don't need to concern yourself with the visual aspect; you're dealing with a text, and PlantUML takes care of the graphical representation for you. You can work with and store simple text files instead of dealing with pictures This is particularly beneficial when encountering images of diagrams in documentation. The quality may be compromised. It's impossible to continue working with such diagrams, and they can only be redrawn, not edited or extended. Another significant advantage is that , allowing you to enjoy the benefits of code generation and seamlessly integrate them with UML. I'll share some examples from my personal experience below. ChatGPT understands PlantUML and can generate code You can use any code editors for PluntUML. I personally prefer Visual Studio Code and PlantUML extension in it. Another way is to use online redactors like or others. https://plantuml.com ChatGPT, Write PluntUML Code for Me Example 1 – I Don’t Wanna Start It With a Blank Slate Let’s create a Use Case diagram for a Taxi application. First, I use a prompt: use syntax from https://plantuml.com/ru/use-case-diagram, create a use case diagram for a taxi app ChatGPT generates a code: @startuml
left to right direction
actor "Passenger" as passenger
actor "Driver" as driver
rectangle "Taxi App" {
  usecase "Request Ride" as request
  usecase "Cancel Ride" as cancel
  usecase "Track Ride" as track
  usecase "Payment" as payment

  passenger -- (request)
  passenger -- (cancel)
  passenger -- (track)
  passenger -- (payment)

  driver -- (track)
  driver -- (payment)

  (request) --> (cancel)
  (request) --> (track)
  (track) --> (payment)
}
@enduml Visual view: The idea is fine, but having only four cases seems insufficient. I request to add additional cases and enhance it with some syntax examples, as there are none similar in PlantUML. Prompt: add more cases, you can use different types of links such as include, extends and generalization, syntax: (case1) .> (case2_includes_case1) : <<include>> (parent) <|-- (child) (case1 extends case 2) .> (case2) : <<extends>> Then I ask multiple times to add more cases again and make some specific requests to fix errors. Here is the final outcome from ChatGPT: @startuml
left to right direction
actor "Passenger" as passenger
actor "Driver" as driver
rectangle "Taxi App" {
  usecase "Request Ride" as request
  usecase "Cancel Ride" as cancel
  usecase "Track Ride" as track
  usecase "Payment" as payment
  usecase "Rate Driver" as rate
  usecase "View Ride History" as history
  usecase "Select Ride Type" as selectType
  usecase "Provide Feedback" as feedback
  usecase "View Driver Information" as driverInfo
  usecase "Update Profile" as updateProfile
  usecase "Promo Code" as promoCode
  usecase "Emergency Assistance" as emergencyAssistance

  passenger -- (selectType)
  passenger -- (request)
  passenger -- (cancel)
  passenger -- (track)
  passenger -- (payment)
  passenger -- (rate)
  passenger -- (history)
  passenger -- (feedback)
  passenger -- (driverInfo)
  passenger -- (updateProfile)
  passenger -- (promoCode)
  passenger -- (emergencyAssistance)

  driver -- (track)
  driver -- (payment)
  driver -- (rate)

  (request) ---> (cancel)
  (request) ---> (track)
  (track) ---> (payment)
  (track) ---> (rate)
  (history) --> (payment)
  (rate) --> (payment)
  (rate) ---> (track)
  (cancel) --> (request)
  (selectType) --> (request)
  (feedback) --> (rate)
  (driverInfo) ---> (track)
  (updateProfile) --> (track)
  (promoCode) --> (payment)
  (emergencyAssistance) --> (track)
}
@enduml I like the cases, but not the connections between them. So, I edit that code a bit. Here is the final result: @startuml

actor "Passenger" as passenger
actor "Driver" as driver
rectangle "Taxi App" {
  usecase "Request Ride" as request
  usecase "Cancel Ride" as cancel
  usecase "Track Ride" as track
  usecase "Payment" as payment
  usecase "Rate Driver" as rate
  usecase "View Ride History" as history
  usecase "Select Ride Type" as selectType
  usecase "Provide Feedback" as feedback
  usecase "View Driver Information" as driverInfo
  usecase "Update Profile" as updateProfile
  usecase "Promo Code" as promoCode
  usecase "Emergency Assistance" as emergencyAssistance

  passenger -- (selectType)
  passenger - (request)
  passenger -- (track)
  passenger -- (payment)
  passenger -- (history)
  passenger -- (rate)
  passenger -- (updateProfile)
  passenger -- (emergencyAssistance)

  driver - (request)
  driver - (payment)
  driver - (rate)
  (request) --> (track)
  (track) --> (payment)
  (track) --> (rate)

  (selectType) --> (request)
  (cancel) .> (request) : <<extends>>
  (rate) <.. (feedback)  : <<include>>
  (payment) <..(promoCode) : <<include>>
  (driverInfo) .> (track) : <<extends>>
}
@enduml Certainly, you can incorporate more scenarios. For example, there's no information about user registration and login. Furthermore, you can guide ChatGPT to create a diagram related to your designs. The key is that you have a foundation that you can use. Example 2 – I Need to Check and Enhance a Diagram I have a simple version of a sequence diagram for picture storage solution with a reverse proxy: @startuml
participant Client
boundary "Reverse Proxy" as Proxy
participant Back
participant FileStorage
participant Admin
actor Editor

Editor -> Admin: Add picture
Admin -> FileStorage: Add picture

Client -> Proxy: Get picture
Proxy -> Proxy: Check cache

alt cash is not empty
Proxy -->> Client: Return picture
else cash is empty
Proxy -> FileStorage: Recieve picture
FileStorage --> Proxy: Picture
Proxy -> Proxy: Save to cache
Proxy -->> Client: Return picture
end
@enduml Let’s improve it. My prompt: I'm designing a solution for storing and using images, check this diagram and edit it if necessary ChatGPT suggestion: 1. **Participant Name Clarification:**
    - Consider renaming "Back" to something more descriptive, like "Application Server" or a similar term that better represents the functionality.
2. **Actor Placement:**
    - Move the "Editor" actor to a more standard position, usually at the top.

@startuml
participant Client
actor Editor
boundary "Reverse Proxy" as Proxy
participant ApplicationServer
participant FileStorage
participant Admin

Editor -> Admin: Add picture
Admin -> FileStorage: Add picture

Client -> Proxy: Get picture
Proxy -> Proxy: Check cache

alt cache is not empty
    Proxy -->> Client: Return picture
else cache is empty
    Proxy -> ApplicationServer: Receive picture
    ApplicationServer --> Proxy: Picture
    Proxy -> Proxy: Save to cache
    Proxy -->> Client: Return picture
end
@enduml I don’t like the point about actor placement, so I put it back; everything else is fine. Then I repeated several times a prompt: check this diagram and enhance it if needed and write prompts: show how ApplicationServer get pic from database add focus of control add some corner cases To finalize it, I adjust the focus of control and make some changes to the naming. Here's the final result: @startuml
participant Client
boundary "Reverse Proxy" as Proxy
participant ApplicationServer
database FileStorage
participant Admin
actor Editor

Editor -> Admin: Add picture
activate Admin
deactivate Admin
Admin -> FileStorage: Add picture
activate FileStorage
deactivate FileStorage
Admin -> ApplicationServer: Store metadata
activate ApplicationServer

Client -> Proxy: Get picture
activate Proxy
Proxy -> Proxy: Check cache

alt cache is not empty
  Proxy -->> Client: Return picture
else cache is empty
  Proxy -> ApplicationServer: Receive picture request
  activate FileStorage
  ApplicationServer -> FileStorage: Retrieve picture from database
  alt picture not found in database
    FileStorage --> ApplicationServer: Picture not found
    deactivate FileStorage
    ApplicationServer -> Proxy: Notify picture not found
    Proxy -->> Client: Picture not found
  else
    FileStorage --> ApplicationServer: Picture data
    deactivate FileStorage
    ApplicationServer -> Proxy: Send picture to Proxy
    Proxy -> Proxy: Save to cache
    Proxy -->> Client: Return picture
  end
end

deactivate Proxy
deactivate ApplicationServer
deactivate Admin
deactivate Editor
@enduml Drawbacks First of all, you have to check everything because "ChatGPT can make mistakes." It is not meant to be a replacement for you but rather a tool to streamline and optimize your work. The most common errors are related to PlantUML syntax. Sometimes, it happens in the middle of a conversation, and I have to fix it and get ChatGPT back on the right track. Second, I use a free version of ChatGPT 3.5, and it has ups and downs. Sometimes, it works worse and gives some random answers for the same prompt. Third, ChatGPT understands some diagrams better than others. For example, it does much better with sequence diagrams than use case diagrams. However, despite these limitations, ChatGPT remains a valuable tool and an excellent assistant capable of increasing your efficiency and saving you time. Conclusion I use ChatGPT on a permanent basis in my work. If you're a manager, analyst, developer, tester, or in any other role, you can make use of UML. If that's the case, consider writing and generating PlantUML code — it will save you time and boost your efficiency. I hope my experience will be useful and inspire you. Thank you!

ChatGPT: Your Time-Saving Companion for UML Diagram Generation

চ্যাটজিপিটি: ইউএমএল ডায়াগ্রাম জেনারেশনের জন্য আপনার সময় বাঁচানোর সঙ্গী

ChatGPT: Ihr zeitsparender Begleiter für die Erstellung von UML-Diagrammen

ChatGPT: su compañero que le permitirá ahorrar tiempo en la generación de diagramas UML

ChatGPT : votre compagnon qui vous fait gagner du temps pour la génération de diagrammes UML

चैटजीपीटी: यूएमएल आरेख निर्माण के लिए आपका समय बचाने वाला साथी

ChatGPT: UML 図生成の時間を節約するコンパニオン

ChatGPT: UML 다이어그램 생성을 위한 시간 절약 도우미

ChatGPT: seu companheiro que economiza tempo para geração de diagramas UML

ChatGPT: ваш помощник для экономии времени при создании UML-диаграмм

ChatGPT: UML Diyagramı Oluşturmada Zaman Kazandıran Yardımcınız

ChatGPT: Người bạn đồng hành tiết kiệm thời gian của bạn để tạo sơ đồ UML

ChatGPT：生成 UML 图的省时伴侣

Using Python to Save a Dataframe into Oracle Database You can use the library in combination with the and libraries to save a dataframe to an Oracle database in Python. pandas sqlalchemy cx_Oracle import pandas as pd
from sqlalchemy import create_engine
import cx_Oracle

# Create an SQLAlchemy engine using the connection
engine = create_engine('oracle://username:password@hostname:port/service_name')

# Save the dataframe to the Oracle database and lowercase your_table_name 
df.to_sql('your_table_name', con=engine, if_exists='replace', index=False) Replace , , , , and with your actual Oracle database credentials and connection details. 'username' 'password' 'hostname' 'port' 'service_name' Please replace with the actual lowercase name of the table in your Oracle database. Otherwise, you may encounter . your_table_name InvalidRequestError: Could not reflect: requested table(s) not available in Engine When you run this script, it will establish a connection to the database, create a sample dataframe, and then save the dataframe to the specified table in the Oracle database. Oracle Make sure the table structure has been created in the database before proceeding. Option 1: Create Table Structure using SQL Keep in mind that you will need to create the table structure in the Oracle database before attempting to save the dataframe to it. The table structure should match the structure of the dataframe in terms of column names and data types. You can create the table structure in your Oracle database using SQL. Here's an example of how you might create a table in Oracle to match the structure of your dataframe: CREATE TABLE your_table_name (
    column1_name  column1_data_type,
    column2_name  column2_data_type,
    ...
); Option 2: Create Table Structure using Python You can run the SQL script in Python using the method provided by the SQLAlchemy library. Here's an example of how you can execute an SQL script after creating the engine: engine.execute() from sqlalchemy import create_engine

# Create an engine
engine = create_engine('oracle://username:password@hostname:port/database_name')

# SQL script to create a table
sql_script =
CREATE TABLE your_table_name (
    column1_name  column1_data_type,
    column2_name  column2_data_type,
    ...
);

# Execute the SQL script
with engine.connect() as connection:
    connection.execute(sql_script) Make sure to replace , , , etc., with your actual table and column names and data types. your_table_name column1_name column1_data_type Conclusion Once the table structure is in place, you can then use the method I mentioned earlier to save the dataframe to the Oracle database. The method will automatically insert the data from the dataframe into the existing table in the database as long as the column names and data types match. to_sql Thank you for taking the time to explore data-related insights with me. I appreciate your engagement. If you find this information helpful, I invite you to follow me or connect with me on . Happy exploring!👋 LinkedIn

Saving Dataframes into Oracle Database with Python

Knowledge base vs LLM What if we could create a digital inventory of all human knowledge, an encyclopedia but ordered in such a way that complex queries can be answered? I can hear you shout: ‘Isn’t that a bit like a LLM trained on Wikipedia?' Kind of … Except that such would have created the inventory itself, leaving the connections between facts opaque. This is so because all text input data fed into an LLM is unstructured and treated on equal footing. It’s the internal attention mechanism that starts organising the text.  So, although we can ask complex queries in the form of natural language and retrieve an answer in the same medium, there is no knowledge inventory accessible to us. It’s as if the LLM is giving a press conference but leaves us dry with ‘no further questions’ after the first round. an LLM Is that better than having no query capability at all? Well, that’s really a mis-apperception of the playing field. The choice isn’t between LLM-like query magic and nothing at all. The choice is between LLM-like query magic and incomplete yet carefully crafted knowledge bases (search engines can answer queries but do so by pointing to documents on the web rather than providing the answer themselves). Using the earlier press conference analogy, the choice is between listening to a spokesman with a vast vocabulary and asking the actual protagonists at the risk of coming home empty-handed. Medical Ontology exposing LLM limits. The foundational model of anatomy (FMA) is such an attempt an create an inventory of human knowledge, in this particular case, all knowledge related to the structural organisation of the human body at micro- and macroscopic levels. A prima facie uncontroversial choice of fundamental relationship to be used in this knowledge base is the partonomic inclusion relationship. For example, . Little spoiler… this slicing up of the conceptual space doesn’t cut it (pun intended). I’ll explain to you why. However, let’s first focus on what this means within the here-discussed wider context. cytoplasm part_of cell We pride ourselves on having LLMs which can assist us in digging into vast amounts of knowledge, conveniently ignoring the fact that we have no real insight into how facts are connected within the model, while our own attempts to connect facts in the most rigorous fashion have turned out to be treacherous endeavours. The press conference gives us the impression of having received some reliable answers while even the protagonists themselves can’t really recall what has happened. If we’re a member of the public relationships committee, this might not worry us too much, but as a citizen with a vested interest in the upholding of liberal democracy, I might not want to rely too heavily on such reports. Partonomic inclusion has gone wrong Now, how can a relationship so seemingly innocent and straightforward as partonomic inclusion be so misleading? Here are some seemingly uncontroversial examples of such a relationship: human testis part_of human being Yes, but there are male and female human beings. So, every human testis belongs (or belonged if the human was subject to an unlikely accident) to a human, but not every human has a human testis. Solution? The part_of relationship is split into two sub-relationships: has_part and part_for. So, now we can state that the human testis part_for human being without committing to human being has_part human testis. The part_of relationship is salvaged as the conjunction of these two sub-relationships: A part_of B =def A part_for B & B has_part A. replication fork part_of nucleoplasm Yes, but only so during certain times during the cell replication cycle (i.e., the S phase). Biology read: https://www.nature.com/scitable/topicpage/replication-fork-stalling-and-the-fork-protection-14435782/ Solution? At first glance, there seem to be at least two possible ways out. Firstly, we could introduce time-dependent or phase-dependent mereological relationships. So, part_of could get a time-index to distinguish between different times at which a relationship is obtained. However, since the cell replication cycle consists of clearly defined phases, it’s logically more apt to introduce phase-dependent partonomic inclusion relationships. Secondly, instead of fiddling with the relationship itself, we can enrich the representational capacity of the relata (i.e., the things related). So, a replication fork is not just a replication fork tout court, nor is the nucleoplasm always the same entity. The partonomic inclusion of the replication fork in the nucleoplasm during the S phase can be represented as follows: replication fork part_of nucleoplasm_S. synaptonemal complex part_of chromosome Yes, but not in all cases. For example, in protozoan ciliates, the formation of a synaptonemal complex is not required for pairing homologous chromosomes. Biology read: https://pubmed.ncbi.nlm.nih.gov/24336924/ Solution? In this case, recourse is found in the ancillary axioms, which define the logical relationships between the fundamental mereological relationships at hand. If we want to represent ‘part of’ as ‘can be part of,‘ then we can introduce a less fundamental part_of_can relationship by defining it as an existential import relationship: A part_of_can B =def C(C is_a B & A part_of C). So, for this particular example, we’d have synaptonemal complex part_of_can chromosome, which is to say that there exists some chromosome for which a synaptonemal complex is part of it. The attentive reader will, at this point, see an alternative solution akin to one earlier exposed. Indeed, instead of introducing a secondary relationship, one can stick to the fundamental relationships and represent the facts by demultiplexing the relata. So, the fact ‘synaptonemal complex part_of chromosome_Animalia’ would be part of the knowledge base but not the fact ‘synaptonemal complex part_of chromosome_Ciliophora.’ ∃ Conclusion In short, prima facia well-rounded relationships need to be broken up and defined in more primitive terms or require ancillary axioms in order to capture the complexity of the reality one desires to represent. Moreover, these solutions are in tension with solutions that prefer relational and definitional parsimony at the cost of an increase in entity cardinality. Since these issues occur in medical ontologies designed by professionals, it should be clear that this task is under no circumstance an easy one. It highlights a perplexing double bind between, on the one hand, the subtlety of reality’s relational composition and, on the other hand, the flexibility of the natural language to squeeze and squash that marvelous detail into communicable trunks. Now, what are the implications of this insight on the state of AI’s contemporary playing field? If we aim for accuracy of representation, there might still be a role for classical AI or resurrections of its core ideas within deep learning paradigms. Conversely, if we aim for smoothness of communication, LLM might actually be capturing exactly what we want… were it not for the fact that we, humans ourselves, have already mastered the skill of human language so that LLM become what sex dolls are in comparison to real intimacy. Sources Paper on which this article is based: Bary Smith and Cornelius Rosse - The role of foundational relations in the alignment of biomedical ontologies: https://pubmed.ncbi.nlm.nih.gov/15360852/

Saving Medical Ontologies with Formal Logic: A Tale of Caution and Hope for Classical AI

On the Economy of Machinery and Manufactures by Charles Babbage is part of the HackerNoon Books Series. You can jump to any chapter in this book . Saving time in Natural Operations here Chapter 6. Saving time in Natural Operations 47. The process of tanning will furnish us with a striking illustration of the power of machinery in accelerating certain processes in which natural operations have a principal effect. The object of this art is to combine a certain principle called tanning with every particle of the skin to be tanned. This, in the ordinary process, is accomplished by allowing the skins to soak in pits containing a solution of tanning matter: they remain in the pits six, twelve, or eighteen months; and in some instances (if the hides are very thick), they are exposed to the operation for two years, or even during a longer period. This length of time is apparently required in order to allow the tanning matter to penetrate into the interior of a thick hide. The improved process consists in placing the hides with the solution of tan in close vessels, and then exhausting the air. The effect is to withdraw any air which may be contained in the pores of the hides, and to aid capillary attraction by the pressure of the atmosphere in forcing the tan into the interior of the skins. The effect of the additional force thus brought into action can be equal only to one atmosphere, but a further improvement has been made: the vessel containing the hides is, after exhaustion, filled up with a solution of tan; a small additional quantity is then injected with a forcing-pump. By these means any degree of pressure may be given which the containing vessel is capable of supporting; and it has been found that, by employing such a method, the thickest hides may be tanned in six weeks or two months. 48. The same process of injection might be applied to impregnate timber with tar, or any other substance capable of preserving it from decay, and if it were not too expensive, the deal floors of houses might thus be impregnated with alumine or other substances, which would render them much less liable to be accidentally set on fire. In some cases it might be useful to impregnate woods with resins, varnish, or oil; and wood saturated with oil might, in some instances, be usefully employed in machinery for giving a constant, but very minute supply of that fluid to iron or steel, against which it is worked. Some idea of the quantity of matter which can be injected into wood by great pressure, may be formed, from considering the fact stated by Mr Scoresby, respecting an accident which occurred to a boat of one of our whaling-ships. The harpoon having been struck into the fish, the whale in this instance, dived directly down, and carried the boat along with him. On returning to the surface the animal was killed, but the boat, instead of rising, was found suspended beneath the whale by the rope of the harpoon; and on drawing it up, every part of the wood was found to be so completely saturated with water as to sink immediately to the bottom. 49. The operation of bleaching linen in the open air is one for which considerable time is necessary; and although it does not require much labour, yet, from the risk of damage and of robbery from long exposure, a mode of shortening the process was highly desirable. The method now practised, although not mechanical, is such a remarkable instance of the application of science to the practical purposes of manufactures, that in mentioning the advantages derived from shortening natural operations, it would have been scarcely pardonable to have omitted all allusion to the beautiful application of chlorine, in combination with lime, to the art of bleaching. 50. Another instance more strictly mechanical occurs in some countries where fuel is expensive, and the heat of the sun is not sufficient to evaporate the water from brine springs. The water is first pumped up to a reservoir, and then allowed to fall in small streams through faggots. Thus it becomes divided; and, presenting a large surface, evaporation is facilitated, and the. brine which is collected in the vessels below the faggots is stronger than that which was pumped up. After thus getting rid of a large part of the water, the remaining portion is driven off by boiling. The success of this process depends on the condition of the atmosphere with respect to moisture. If the air, at the time the brine falls through the faggots, holds in solution as much moisture as it can contain in an invisible state, no more can be absorbed from the salt water, and the labour expended in pumping is entirely wasted. The state of the air, as to dryness, is therefore an important consideration in fixing the time when this operation is to be performed; and an attentive examination of its state, by means of the hygrometer, might be productive of some economy of labour. 51. In some countries, where wood is scarce, the evaporation of salt water is carried on by a large collection of ropes which are stretched perpendicularly. In passing down the ropes, the water deposits the sulphate of lime which it held in solution, and gradually incrusts them, so that in the course of twenty years, when they are nearly rotten, they are still sustained by the surrounding incrustation, thus presenting the appearance of a vast collection of small columns. 52. Amongst natural operations perpetually altering the surface of our globe, there are some which it would be advantageous to accelerate. The wearing down of the rocks which impede the rapids of navigable rivers, is one of this class. A very beautiful process for accomplishing this object has been employed in America. A boat is placed at the bottom of the rapid, and kept in its position by a long rope which is firmly fixed on the bank of the river near the top. An axis, having a wheel similar to the paddle-wheel of a steamboat fixed at each end of it, is placed across the boat; so that the two wheels and their connecting axis shall revolve rapidly, being driven by the force of the passing current. Let us now imagine several beams of wood shod with pointed iron fixed at the ends of strong levers, projecting beyond the bow of the boat, as in the annexed representation. If these levers are at liberty to move up and down, and if one or more projecting pieces, called cams, are fixed on the axis opposite to the end of each lever, the action of the stream upon the wheels will keep up a perpetual succession of blows. The sharp-pointed shoe striking upon the rock at the bottom, will continually detach small pieces, which the stream will immediately carry off. Thus, by the mere action of the river itself, a constant and most effectual system of pounding the rock at its bottom is established. A single workman may, by the aid of a rudder, direct the boat to any required part of the stream; and when it is necessary to move up the rapid, as the channel is cut, he can easily cause the boat to advance by means of a capstan. 53. When the object of the machinery just described has been accomplished, and the channel is sufficiently deep, a slight alteration converts the apparatus to another purpose almost equally advantageous. The stampers and the projecting pieces on the axis are removed, and a barrel of wood or metal, surrounding part of the axis, and capable, at pleasure, of being connected with, or disconnected from the axis itself, is substituted. The rope which hitherto fastened the boat, is now fixed to this barrel; and if the barrel is loose upon the axis, the paddle-wheel makes the axis only revolve, and the boat remains in its place: but the moment the axis is attached to its surrounding barrel, this begins to turn, and winding up the rope, the boat is gradually drawn up against the stream; and may be employed as a kind of tug-boat for vessels which have occasion to ascend the rapid. When the tug-boat reaches the summit the barrel is released from the axis, and friction being applied to moderate its velocity, the boat is allowed to descend. 54. Clocks occupy a very high place amongst instruments by means of which human time is economized: and their multiplication in conspicuous places in large towns is attended with many advantages. Their position, nevertheless, in London, is often very ill chosen; and the usual place, halfway up on a high steeple, in the midst of narrow streets, in a crowded city, is very unfavourable, unless the church happen to stand out from the houses which form the street. The most eligible situation for a clock is, that it should project considerably into the street at some elevation, with a dial-plate on each side, like that which belonged to the old church of St Dunstan, in Fleet Street, so that passengers in both directions would have their attention directed to the hour. 55. A similar remark applies, with much greater force, to the present defective mode of informing the public of the position of the receiving houses for the two penny and general post. In the lowest corner of the window of some attractive shop is found a small slit, with a brass plate indicating its important office so obscurely that it seems to be an object rather to prevent its being conspicuous. No striking sign assists the anxious enquirer, who, as the moments rapidly pass which precede the hour of closing, torments the passenger with his enquiries for the nearest post-office. He reaches it, perhaps, just as it is closed; and must then either hasten to a distant part of the town in order to procure the admission of his letters or give up the idea of forwarding them by that post; and thus, if they are foreign letters, he may lose, perhaps, a week or a fortnight by waiting for the next packet. The inconvenience in this and in some other cases, is of perpetual and everyday occurrence; and though, in the greater part of the individual cases, it may be of trifling moment, the sum of all these produces an amount, which it is always worthy of the government of a large and active population to attend to. The remedy is simple and obvious: it would only be necessary, at each letter-box, to have a light frame of iron projecting from the house over the pavement, and carrying the letters G. P., or T. P., or any other distinctive sign. All private signs are at present very properly prohibited from projecting into the street: the passenger, therefore, would at once know where to direct his attention, in order to discover a post-office; and those letter-boxes which occurred in the great thoroughfares could not fail to be generally known. About HackerNoon Book Series: We bring you the most important technical, scientific, and insightful public domain books. This book is part of the public domain. Charles Babbage (2003). On the Economy of Machinery and Manufactures. Urbana, Illinois: Project Gutenberg. Retrieved https://www.gutenberg.org/cache/epub/4238/pg4238.html This eBook is for the use of anyone anywhere at no cost and with almost no restrictions whatsoever. You may copy it, give it away or re-use it under the terms of the Project Gutenberg License included with this eBook or online at , located at . www.gutenberg.org https://www.gutenberg.org/policy/license.html

Saving time in Natural Operations

Saving Time and Nerves With Formulas and the Structure Jira Plugin

সূত্র এবং গঠন জিরা প্লাগইন সহ সময় এবং স্নায়ু সংরক্ষণ

Zeit und Nerven sparen mit Formeln und dem Structure Jira Plugin

Ahorrando tiempo y nervios con fórmulas y el complemento Structure Jira

Gagner du temps et des nerfs avec les formules et le plugin Structure Jira

फ़ॉर्मूले और संरचना जीरा प्लगइन के साथ समय और घबराहट की बचत

数式と構造 Jira プラグインで時間と神経を節約

수식과 구조 Jira 플러그인을 사용하여 시간과 신경을 절약하세요

Economizando tempo e nervosismo com fórmulas e estrutura Jira Plugin

Экономия времени и нервов с помощью формул и плагина Jira Structure

Formüller ve Yapıyla Zamandan ve Sinirlerden Tasarruf Edin Jira Eklentisi

Tiết kiệm thời gian và thần kinh với các công thức và Plugin Jira Cấu trúc

使用公式和结构 Jira 插件节省时间和精力

Introducing PromptCompressor, a web service that a cost-effective way to use language models like ChatGPT. PromptCompressor employs a deep learning model trained through reinforcement learning to reduce the length of the original prompts by approximately 20-30% while maintaining their performance. offers Getting Started with PromptCompressor Enter your prompts into the text box at . Copy the compressed prompt and input it into ChatGPT. promptcompressor.com The compressed prompts may contain many grammatical errors but still work effectively. Sometimes, they even outperform! PromptCompressor has been trained on the following prompt format. For stable performance, please write your prompts according to this format: Instruction: <Write the instruction here. For example, 'Summarize the key points of the text.'> Input: <Here, enter the sentence or data to be processed according to the instruction. Maximize Your Tokens, Minimize Your Costs If you are providing a service using the GPT API or using it personally, you might have encountered issues with API costs or the constraints of the context window. The current pricing policy and maximum context size of GPT are as : follows Model Input Output GPT-3.5-turbo 4K context $0.0015 / 1K tokens $0.002 / 1K tokens 16K context $0.003 / 1K tokens $0.004 / 1K tokens GPT-4 8K context $0.03 / 1K tokens $0.06 / 1K tokens 32K context $0.06 / 1K tokens $0.12 / 1K tokens In simple terms: Tokens are the smallest units that need to be processed when analyzing or generating text. In English, a word, symbol, or space can be a token. Typically, 1,000 tokens can represent about 750 words. Context is the maximum number of tokens that an AI model can process at once. For example, a 4K context means the model can process up to 4,000 tokens at a time. The total cost is calculated by multiplying the cost per token for both input and output tokens. Therefore, PromptCompressor is suitable for tasks like summarization, prediction, classification, question answering, where the input tokens are more than the output tokens. How Does It Work? Humans can understand the meaning of a sentence even if some words are missing, such as articles like "a" and "the", prepositions like "in" and "of", etc. GPT, trained on human language, works similarly. We used this idea to train a model that removes tokens that have less impact on GPT's output, reducing the length of the prompt while retaining the meaning and context. Evaluating the Performance We conducted experiments with over 100,000 instructions on the model . We confirmed that it maintains about 95% of the performance while achieving an average compression effect of about 1/4. Llama-2-7b-chat-hf PromptCompressor generates the optimal prompts considering the context. Therefore, the compression rate can vary depending on the task. For instance, compression may not be performed in cases where there are hardly any unnecessary tokens to eliminate. What’s Next Currently, PromptCompressor has a token limit of 507 due to the limitations of the backbone model. We aim to increase this limit and are continuously researching and developing to improve the service. Contact If you have questions or feedback about the service, please contact . Your opinions will be a great help in improving the service. ghdbsl98@gmail.com

PromptCompressor: Enjoy 25% Savings on GPT API Without Losing Quality!

Stories, News & Mentions about SAVI

Savi’s Stories on HackerNoon