<![CDATA[Vitor Pamplona]]> https://vitor.npub.pro/tag/nostr/ https://vitor.npub.pro/tag/nostr/rss/ episodic Wed, 09 Oct 2024 17:46:49 GMT Wed, 09 Oct 2024 17:46:49 GMT <![CDATA[Vitor Pamplona]]> https://vitor.npub.pro/tag/nostr/ https://vitorpamplona.com/me.jpg <![CDATA[Does Size Matter? How big is too big?]]> Wed, 09 Oct 2024 17:46:49 GMT https://vitor.npub.pro/post/1728488779653/ https://vitor.npub.pro/post/1728488779653/ naddr1qqxnzdej8q6rswphxuunvdfnqgsyvrp9u6p0mfur9dfdru3d853tx9mdjuhkphxuxgfwmryja7zsvhqrqsqqqa28h5f6jy Nostr naddr1qqxnzdej8q6rswphxuunvdfnqgsyvrp9u6p0mfur9dfdru3d853tx9mdjuhkphxuxgfwmryja7zsvhqrqsqqqa28h5f6jy npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z In the early days of Nostr, developers often competed to see who could implement the most NIPs. Although all were optional (except NIP-01), it became a point of pride and vital for the ecosystem's growth. Back then, there were only a few dozen relatively simple NIPs to implement. Fast forward to today, with nearly 100 NIPs, maintaining and implementing everything has become nearly impossible. Yet, the drive among developers to "code all things Nostr" remains as strong as ever.

fiatjaf raised the point that everyone, even I, agrees:

nostr:nevent1qqsqqqp2zrs7836tyjlsfe7aj9c4d97zrxxqyayagkdwlcur96t4laspzemhxue69uhhyetvv9ujumt0wd68ytnsw43z7q3q80cvv07tjdrrgpa0j7j7tmnyl2yr6yr7l8j4s3evf6u64th6gkwsxpqqqqqqzgcrrrp

But how big is too big? How can we better understand the range of options available for devs out there?

I went out for a hunt in my own brain to figure out how to clarify the situation. I came up with the following 4 categories for Nostr Clients:

  • Super Clients: These apps merge vastly different domains into a single application, offering basic support for reading, writing, configuration, and data management for each use case within each domains. An example would be an app that combines a Marketplace and Live Streams under one roof.

  • Clients: These apps provide comprehensive support for a single domain, handling all its use cases in a single home. They manage the complete set of reading, writing, configuration, and long-term data management within that domain. An example is a marketplace app that helps users manage product catalogs, process orders, collect payments, and handle fulfillment and reports.

  • Mini Clients: These apps focus on read and write functionality for a single use case, including configuration management and any actions related to that specific task. For example, a fulfillment app that helps users view orders placed from another client to then pack and ship them.

  • Micro Clients: These apps have a single interface and perform one specific action. Viewing and creating a record is handled by separate micro apps. An example is an app that simply scans an order's QR code and marks it as shipped.

Based on my made-up categories described at the end, this is how I would split our most known apps.

Super Clients

Clients

Mini Clients

Micro Clients

Super apps will try to do everything, but can't really do most things super well. Regular-sized Clients will try to manage most of a given domain but are likely to centralize users on themselves, an unwanted effect inside of Nostr. If we want Nostr to grow in a decentralized fashion, we have to start betting on and using more Mini and Micro clients.

]]> In the early days of Nostr, developers often competed to see who could implement the most NIPs. Although all were optional (except NIP-01), it became a point of pride and vital for the ecosystem's growth. Back then, there were only a few dozen relatively simple NIPs to implement. Fast forward to today, with nearly 100 NIPs, maintaining and implementing everything has become nearly impossible. Yet, the drive among developers to "code all things Nostr" remains as strong as ever.

fiatjaf raised the point that everyone, even I, agrees:

nostr:nevent1qqsqqqp2zrs7836tyjlsfe7aj9c4d97zrxxqyayagkdwlcur96t4laspzemhxue69uhhyetvv9ujumt0wd68ytnsw43z7q3q80cvv07tjdrrgpa0j7j7tmnyl2yr6yr7l8j4s3evf6u64th6gkwsxpqqqqqqzgcrrrp

But how big is too big? How can we better understand the range of options available for devs out there?

I went out for a hunt in my own brain to figure out how to clarify the situation. I came up with the following 4 categories for Nostr Clients:

  • Super Clients: These apps merge vastly different domains into a single application, offering basic support for reading, writing, configuration, and data management for each use case within each domains. An example would be an app that combines a Marketplace and Live Streams under one roof.

  • Clients: These apps provide comprehensive support for a single domain, handling all its use cases in a single home. They manage the complete set of reading, writing, configuration, and long-term data management within that domain. An example is a marketplace app that helps users manage product catalogs, process orders, collect payments, and handle fulfillment and reports.

  • Mini Clients: These apps focus on read and write functionality for a single use case, including configuration management and any actions related to that specific task. For example, a fulfillment app that helps users view orders placed from another client to then pack and ship them.

  • Micro Clients: These apps have a single interface and perform one specific action. Viewing and creating a record is handled by separate micro apps. An example is an app that simply scans an order's QR code and marks it as shipped.

Based on my made-up categories described at the end, this is how I would split our most known apps.

Super Clients

Clients

Mini Clients

Micro Clients

Super apps will try to do everything, but can't really do most things super well. Regular-sized Clients will try to manage most of a given domain but are likely to centralize users on themselves, an unwanted effect inside of Nostr. If we want Nostr to grow in a decentralized fashion, we have to start betting on and using more Mini and Micro clients.

]]> <![CDATA[Relay Setup 101]]> Thu, 29 Aug 2024 00:49:42 GMT https://vitor.npub.pro/post/relay-setup/ https://vitor.npub.pro/post/relay-setup/ naddr1qq9hyetvv9uj6um9w36hqq3qgcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqxpqqqp65wjvcq4q Amethyst naddr1qq9hyetvv9uj6um9w36hqq3qgcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqxpqqqp65wjvcq4q npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z If you don't know where your posts are, you might as well just stay in the centralized Twitter. You either take control of your relay lists, or they will control you. Amethyst offers several lists of relays for our users. We are going to go one by one to help clarify what they are and which options are best for each one.

Public Home/Outbox Relays

Home relays store all YOUR content: all your posts, likes, replies, lists, etc. It's your home. Amethyst will send your posts here first. Your followers will use these relays to get new posts from you. So, if you don't have anything there, they will not receive your updates.

Home relays must allow queries from anyone, ideally without the need to authenticate. They can limit writes to paid users without affecting anyone's experience.

This list should have a maximum of 3 relays. More than that will only make your followers waste their mobile data getting your posts. Keep it simple. Out of the 3 relays, I recommend:

  • 1 large public, international relay: nos.lol, nostr.mom, relay.damus.io, etc.
  • 1 personal relay to store a copy of all your content in a place no one can delete. Go to relay.tools and never be censored again.
  • 1 really fast relay located in your country: paid options like http://nostr.wine are great

Do not include relays that block users from seeing posts in this list. If you do, no one will see your posts.

Public Inbox Relays

This relay type receives all replies, comments, likes, and zaps to your posts. If you are not getting notifications or you don't see replies from your friends, it is likely because you don't have the right setup here. If you are getting too much spam in your replies, it's probably because your inbox relays are not protecting you enough. Paid relays can filter inbox spam out.

Inbox relays must allow anyone to write into them. It's the opposite of the outbox relay. They can limit who can download the posts to their paid subscribers without affecting anyone's experience.

This list should have a maximum of 3 relays as well. Again, keep it small. More than that will just make you spend more of your data plan downloading the same notifications from all these different servers. Out of the 3 relays, I recommend:

  • 1 large public, international relay: nos.lol, nostr.mom, relay.damus.io, etc.
  • 1 personal relay to store a copy of your notifications, invites, cashu tokens and zaps.
  • 1 really fast relay located in your country: go to nostr.watch and find relays in your country

Terrible options include:

  • nostr.wine should not be here.
  • filter.nostr.wine should not be here.
  • inbox.nostr.wine should not be here.

DM Inbox Relays

These are the relays used to receive DMs and private content. Others will use these relays to send DMs to you. If you don't have it setup, you will miss DMs. DM Inbox relays should accept any message from anyone, but only allow you to download them.

Generally speaking, you only need 3 for reliability. One of them should be a personal relay to make sure you have a copy of all your messages. The others can be open if you want push notifications or closed if you want full privacy.

Good options are:

  • inbox.nostr.wine and auth.nostr1.com: anyone can send messages and only you can download. Not even our push notification server has access to them to notify you.
  • a personal relay to make sure no one can censor you. Advanced settings on personal relays can also store your DMs privately. Talk to your relay operator for more details.
  • a hidden, but public relay if you want DM notifications from our servers.

Make sure to add at least one public relay if you want to see DM notifications.

Private Home Relays

Private Relays are for things no one should see, like your drafts, lists, app settings, bookmarks etc. Ideally, these relays are either local or require authentication before posting AND downloading each user's content. There are no dedicated relays for this category yet, so I would use a local relay like Citrine on Android and a personal relay on relay.tools.

Keep in mind that if you choose a local relay only, a client on the desktop might not be able to see the drafts from clients on mobile and vice versa.

Search relays:

This is the list of relays to use on Amethyst's search and user tagging with @. Tagging and searching will not work if there is nothing here.. This option requires NIP-50 compliance from each relay. Hit the Default button to use all available options on existence today:

  • nostr.wine
  • relay.nostr.band
  • relay.noswhere.com

Local Relays:

This is your local storage. Everything will load faster if it comes from this relay. You should install Citrine on Android and write ws://localhost:4869 in this option.

General Relays:

This section contains the default relays used to download content from your follows. Notice how you can activate and deactivate the Home, Messages (old-style DMs), Chat (public chats), and Global options in each.

Keep 5-6 large relays on this list and activate them for as many categories (Home, Messages (old-style DMs), Chat, and Global) as possible.

Amethyst will provide additional recommendations to this list from your follows with information on which of your follows might need the additional relay in your list. Add them if you feel like you are missing their posts or if it is just taking too long to load them.

My setup

Here's what I use:

  1. Go to relay.tools and create a relay for yourself.
  2. Go to nostr.wine and pay for their subscription.
  3. Go to inbox.nostr.wine and pay for their subscription.
  4. Go to nostr.watch and find a good relay in your country.
  5. Download Citrine to your phone.

Then, on your relay lists, put:

Public Home/Outbox Relays:

  • nostr.wine
  • nos.lol or an in-country relay.
  • <your.relay>.nostr1.com

Public Inbox Relays

  • nos.lol or an in-country relay
  • <your.relay>.nostr1.com

DM Inbox Relays

  • inbox.nostr.wine
  • <your.relay>.nostr1.com

Private Home Relays

  • ws://localhost:4869 (Citrine)
  • <your.relay>.nostr1.com (if you want)

Search Relays

  • nostr.wine
  • relay.nostr.band
  • relay.noswhere.com

Local Relays

  • ws://localhost:4869 (Citrine)

General Relays

  • nos.lol
  • relay.damus.io
  • relay.primal.net
  • nostr.mom

And a few of the recommended relays from Amethyst.

Final Considerations

Remember, relays can see what your Nostr client is requesting and downloading at all times. They can track what you see and see what you like. They can sell that information to the highest bidder, they can delete your content or content that a sponsor asked them to delete (like a negative review for instance) and they can censor you in any way they see fit. Before using any random free relay out there, make sure you trust its operator and you know its terms of service and privacy policies.

]]> If you don't know where your posts are, you might as well just stay in the centralized Twitter. You either take control of your relay lists, or they will control you. Amethyst offers several lists of relays for our users. We are going to go one by one to help clarify what they are and which options are best for each one.

Public Home/Outbox Relays

Home relays store all YOUR content: all your posts, likes, replies, lists, etc. It's your home. Amethyst will send your posts here first. Your followers will use these relays to get new posts from you. So, if you don't have anything there, they will not receive your updates.

Home relays must allow queries from anyone, ideally without the need to authenticate. They can limit writes to paid users without affecting anyone's experience.

This list should have a maximum of 3 relays. More than that will only make your followers waste their mobile data getting your posts. Keep it simple. Out of the 3 relays, I recommend:

  • 1 large public, international relay: nos.lol, nostr.mom, relay.damus.io, etc.
  • 1 personal relay to store a copy of all your content in a place no one can delete. Go to relay.tools and never be censored again.
  • 1 really fast relay located in your country: paid options like http://nostr.wine are great

Do not include relays that block users from seeing posts in this list. If you do, no one will see your posts.

Public Inbox Relays

This relay type receives all replies, comments, likes, and zaps to your posts. If you are not getting notifications or you don't see replies from your friends, it is likely because you don't have the right setup here. If you are getting too much spam in your replies, it's probably because your inbox relays are not protecting you enough. Paid relays can filter inbox spam out.

Inbox relays must allow anyone to write into them. It's the opposite of the outbox relay. They can limit who can download the posts to their paid subscribers without affecting anyone's experience.

This list should have a maximum of 3 relays as well. Again, keep it small. More than that will just make you spend more of your data plan downloading the same notifications from all these different servers. Out of the 3 relays, I recommend:

  • 1 large public, international relay: nos.lol, nostr.mom, relay.damus.io, etc.
  • 1 personal relay to store a copy of your notifications, invites, cashu tokens and zaps.
  • 1 really fast relay located in your country: go to nostr.watch and find relays in your country

Terrible options include:

  • nostr.wine should not be here.
  • filter.nostr.wine should not be here.
  • inbox.nostr.wine should not be here.

DM Inbox Relays

These are the relays used to receive DMs and private content. Others will use these relays to send DMs to you. If you don't have it setup, you will miss DMs. DM Inbox relays should accept any message from anyone, but only allow you to download them.

Generally speaking, you only need 3 for reliability. One of them should be a personal relay to make sure you have a copy of all your messages. The others can be open if you want push notifications or closed if you want full privacy.

Good options are:

  • inbox.nostr.wine and auth.nostr1.com: anyone can send messages and only you can download. Not even our push notification server has access to them to notify you.
  • a personal relay to make sure no one can censor you. Advanced settings on personal relays can also store your DMs privately. Talk to your relay operator for more details.
  • a hidden, but public relay if you want DM notifications from our servers.

Make sure to add at least one public relay if you want to see DM notifications.

Private Home Relays

Private Relays are for things no one should see, like your drafts, lists, app settings, bookmarks etc. Ideally, these relays are either local or require authentication before posting AND downloading each user's content. There are no dedicated relays for this category yet, so I would use a local relay like Citrine on Android and a personal relay on relay.tools.

Keep in mind that if you choose a local relay only, a client on the desktop might not be able to see the drafts from clients on mobile and vice versa.

Search relays:

This is the list of relays to use on Amethyst's search and user tagging with @. Tagging and searching will not work if there is nothing here.. This option requires NIP-50 compliance from each relay. Hit the Default button to use all available options on existence today:

  • nostr.wine
  • relay.nostr.band
  • relay.noswhere.com

Local Relays:

This is your local storage. Everything will load faster if it comes from this relay. You should install Citrine on Android and write ws://localhost:4869 in this option.

General Relays:

This section contains the default relays used to download content from your follows. Notice how you can activate and deactivate the Home, Messages (old-style DMs), Chat (public chats), and Global options in each.

Keep 5-6 large relays on this list and activate them for as many categories (Home, Messages (old-style DMs), Chat, and Global) as possible.

Amethyst will provide additional recommendations to this list from your follows with information on which of your follows might need the additional relay in your list. Add them if you feel like you are missing their posts or if it is just taking too long to load them.

My setup

Here's what I use:

  1. Go to relay.tools and create a relay for yourself.
  2. Go to nostr.wine and pay for their subscription.
  3. Go to inbox.nostr.wine and pay for their subscription.
  4. Go to nostr.watch and find a good relay in your country.
  5. Download Citrine to your phone.

Then, on your relay lists, put:

Public Home/Outbox Relays:

  • nostr.wine
  • nos.lol or an in-country relay.
  • <your.relay>.nostr1.com

Public Inbox Relays

  • nos.lol or an in-country relay
  • <your.relay>.nostr1.com

DM Inbox Relays

  • inbox.nostr.wine
  • <your.relay>.nostr1.com

Private Home Relays

  • ws://localhost:4869 (Citrine)
  • <your.relay>.nostr1.com (if you want)

Search Relays

  • nostr.wine
  • relay.nostr.band
  • relay.noswhere.com

Local Relays

  • ws://localhost:4869 (Citrine)

General Relays

  • nos.lol
  • relay.damus.io
  • relay.primal.net
  • nostr.mom

And a few of the recommended relays from Amethyst.

Final Considerations

Remember, relays can see what your Nostr client is requesting and downloading at all times. They can track what you see and see what you like. They can sell that information to the highest bidder, they can delete your content or content that a sponsor asked them to delete (like a negative review for instance) and they can censor you in any way they see fit. Before using any random free relay out there, make sure you trust its operator and you know its terms of service and privacy policies.

]]> <![CDATA[Zap-share slavery and the incentive war]]> Sun, 10 Dec 2023 20:09:35 GMT https://vitor.npub.pro/post/rev-share-slavery/ https://vitor.npub.pro/post/rev-share-slavery/ naddr1qqghyetk94eksctjv5khxmrpwejhy7gzyprqcf0xst760qet2tglytfay2e3wmvh9asdehpjztkceyh0s5r9cqcyqqq823c33m8ht Nostr naddr1qqghyetk94eksctjv5khxmrpwejhy7gzyprqcf0xst760qet2tglytfay2e3wmvh9asdehpjztkceyh0s5r9cqcyqqq823c33m8ht npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z Sending a percentage of your zaps to head straight to your Client's development team, optionally or forcibly, sounds like a great way to sustain a project. And, depending on the team's career goals, it is. However, I am here to argue that this revenue model doesn't serve us well.

Let's be real: if you are developing a client and seeking venture capital, then yes, rev-share is by far your best shot at winning. It's a hot favorite in venture pitches because it hands the company (and its investors) control over the entire field while raking in maximum revenue from all participants. Your performance metrics are very well established, incentives are aligned with the fast-scaling needs, the path to growth is very common and many, many investors know the territory extremely well and will be more willing to easily give you money. If that is your case, go for the usual route of YC -> VC -> PE -> IPO. You won't control anything in the end, but it will make you, and your investors, a lot of money.

That being said, I believe this revenue model bakes in the wrong incentives, and forward-thinking clients should seek alternative revenue-generation strategies.

Storytime: Over the past 15 years, I led a medical device business that integrated a rev-share component. Doctors purchasing our devices gained access to our medical record system. The revenue model for that system was rev-share: the system itself was essentially free, and each time providers submitted a prescription, we earned $0.50. Initially, this model thrived because it scaled with our customer's needs. If they were small businesses, their payments would stay small, proportional to their bumpy use. $0.50/patient was something everyone could afford out of their insurance payouts. It was great for beginners, great for large players, and great for us. Things looked great. But little did I know what it would do to the business. Gradually, our focus shifted from developing devices and enhancing the system to obsessing over increasing the number of prescriptions. We ended up staying quite small as a business, but even at that small size, I could already see the damage the revenue model could make at scale. If we grew, it would be all about how to get providers to file as many prescriptions as possible, even for patients who didn't need one. Controlling the medium upon which customers file prescriptions gives the company immense power to incentivize behaviors that will go against the well-being of the community.

Similar tendencies are visible in other platforms like Substack nowadays. You can't visit your favorite author without being constantly "reminded" to subscribe to other creators. It doesn't matter if you care or not about those topics. What matters is that you subscribe to them. The company doesn't care if you end up never reading their content. That's not their problem. Their goal is to push as much content down your throat as possible so that they can collect little fees here and there. The bigger the amount of content you subscribe (not the quality), the bigger the fees they get.

The same is happening with Lightning wallets and many Bitcoin products that sustain themselves by taking a share of your transactions. They are incentivized to make users spend and not to help them manage their finances well.

Nostr won't be different. Zap-share Clients have an incentive to make you spend. In the long run, that is all they will care about. It doesn't matter if the content is good or not.

You become a Foie gras goose.

Social media has grappled with broken incentives since its inception. If we want to design a better social environment, it's our duty to find business models that are more aligned with the well-being of our users. We don't have a clear answer for that right now, but there is hope: we know what we want. We want to align the size of the user's contributions with products that make users feel better and more productive.

In the long run, this desire will split Clients between creator tools and consumer tools. Creator tools should help authors optimize their revenue stream while consumer tools should help readers optimize for their needs and well-being, independent of how much creator tools try to push content down your throat.

If you are a consumer of content (or more precisely, WHEN you are consuming content), you should reward those clients that hit the sweet spot for you without forcing you into spending more than what's needed. If you are a creator (or more precisely, WHEN you are creating content), you should reward clients who help you optimize the outcomes you desire for your content.

I find it hard to believe a single client can play both sides well. In a decentralized environment like ours, specialized clients will always win.

Remember to choose clients that are on your side. A creator's side is not always aligned with the consumer's side. When the eagles are silent, the parrots begin to jabber.

]]> Sending a percentage of your zaps to head straight to your Client's development team, optionally or forcibly, sounds like a great way to sustain a project. And, depending on the team's career goals, it is. However, I am here to argue that this revenue model doesn't serve us well.

Let's be real: if you are developing a client and seeking venture capital, then yes, rev-share is by far your best shot at winning. It's a hot favorite in venture pitches because it hands the company (and its investors) control over the entire field while raking in maximum revenue from all participants. Your performance metrics are very well established, incentives are aligned with the fast-scaling needs, the path to growth is very common and many, many investors know the territory extremely well and will be more willing to easily give you money. If that is your case, go for the usual route of YC -> VC -> PE -> IPO. You won't control anything in the end, but it will make you, and your investors, a lot of money.

That being said, I believe this revenue model bakes in the wrong incentives, and forward-thinking clients should seek alternative revenue-generation strategies.

Storytime: Over the past 15 years, I led a medical device business that integrated a rev-share component. Doctors purchasing our devices gained access to our medical record system. The revenue model for that system was rev-share: the system itself was essentially free, and each time providers submitted a prescription, we earned $0.50. Initially, this model thrived because it scaled with our customer's needs. If they were small businesses, their payments would stay small, proportional to their bumpy use. $0.50/patient was something everyone could afford out of their insurance payouts. It was great for beginners, great for large players, and great for us. Things looked great. But little did I know what it would do to the business. Gradually, our focus shifted from developing devices and enhancing the system to obsessing over increasing the number of prescriptions. We ended up staying quite small as a business, but even at that small size, I could already see the damage the revenue model could make at scale. If we grew, it would be all about how to get providers to file as many prescriptions as possible, even for patients who didn't need one. Controlling the medium upon which customers file prescriptions gives the company immense power to incentivize behaviors that will go against the well-being of the community.

Similar tendencies are visible in other platforms like Substack nowadays. You can't visit your favorite author without being constantly "reminded" to subscribe to other creators. It doesn't matter if you care or not about those topics. What matters is that you subscribe to them. The company doesn't care if you end up never reading their content. That's not their problem. Their goal is to push as much content down your throat as possible so that they can collect little fees here and there. The bigger the amount of content you subscribe (not the quality), the bigger the fees they get.

The same is happening with Lightning wallets and many Bitcoin products that sustain themselves by taking a share of your transactions. They are incentivized to make users spend and not to help them manage their finances well.

Nostr won't be different. Zap-share Clients have an incentive to make you spend. In the long run, that is all they will care about. It doesn't matter if the content is good or not.

You become a Foie gras goose.

Social media has grappled with broken incentives since its inception. If we want to design a better social environment, it's our duty to find business models that are more aligned with the well-being of our users. We don't have a clear answer for that right now, but there is hope: we know what we want. We want to align the size of the user's contributions with products that make users feel better and more productive.

In the long run, this desire will split Clients between creator tools and consumer tools. Creator tools should help authors optimize their revenue stream while consumer tools should help readers optimize for their needs and well-being, independent of how much creator tools try to push content down your throat.

If you are a consumer of content (or more precisely, WHEN you are consuming content), you should reward those clients that hit the sweet spot for you without forcing you into spending more than what's needed. If you are a creator (or more precisely, WHEN you are creating content), you should reward clients who help you optimize the outcomes you desire for your content.

I find it hard to believe a single client can play both sides well. In a decentralized environment like ours, specialized clients will always win.

Remember to choose clients that are on your side. A creator's side is not always aligned with the consumer's side. When the eagles are silent, the parrots begin to jabber.

]]>