Scratchedia - The Internet on Scratch, that's safe (Applications Closed)

IGABMS

xMysticalCoder wrote:
IGABMS wrote:
IGABMS wrote:
Why are we hosting this on another site?

You guys do realize we're like 100x more likely to accidentially DDOS our networks on accident.
*if we host it on our own devices.

Guys- has anybody else been having issues with getting notified of new posts?
Scratch is always like that. Most of the time i get notifs after a page has passed, but if your lucky you can instantly get a notif. I don't know if there is a specific rule of when you get the notif, or if its random.

Alright thanks!

101Corp

Yes

I like scratchedia

Last edited by 101Corp (May 16, 2021 16:08:37)

This signature is sponsored by RAID: Shadow Legends!
Get 3000 coins and a new fighter by entering the code 101CORP!

(plz don’t download it hurts my head)

kkidslogin

I am not going to join the collab, but here is an idea that I have tested and I know works in scratch 3:

If you have an extra computer lying around (A cheap raspberry pi 4 2GB will do), you can have it run the project continually in a “server mode”. A project in “client mode” send events over the cloud variables, and the server responds with data, possibly in steps. More after lunch

My siggy doesn't fit on ur screen! Select it and use CTRL (Or COMMAND if MAC) + SHIFT + DOWN to see the rest please.

when green flag clicked
forever
write code
repeat until <(bugs) = [0]>
test
fix bugs
end
release
end

Newest songs:
Hack Attack
Surrender
Deaf (NEW!)

My collaboration:
Starsoft Incorporated

What I am:
A proud BBcode user

Windows95 gamer on Windows XP tower,
Windows98 and below,
DOS,
8 bit guy,
Commodore 64,
Commodore- everything,
Z80,
IBM PC
and clones lover.

kkidslogin

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).

My siggy doesn't fit on ur screen! Select it and use CTRL (Or COMMAND if MAC) + SHIFT + DOWN to see the rest please.

when green flag clicked
forever
write code
repeat until <(bugs) = [0]>
test
fix bugs
end
release
end

Newest songs:
Hack Attack
Surrender
Deaf (NEW!)

My collaboration:
Starsoft Incorporated

What I am:
A proud BBcode user

Windows95 gamer on Windows XP tower,
Windows98 and below,
DOS,
8 bit guy,
Commodore 64,
Commodore- everything,
Z80,
IBM PC
and clones lover.

kkidslogin

*And live stuff gets even laggier.

My siggy doesn't fit on ur screen! Select it and use CTRL (Or COMMAND if MAC) + SHIFT + DOWN to see the rest please.

when green flag clicked
forever
write code
repeat until <(bugs) = [0]>
test
fix bugs
end
release
end

Newest songs:
Hack Attack
Surrender
Deaf (NEW!)

My collaboration:
Starsoft Incorporated

What I am:
A proud BBcode user

Windows95 gamer on Windows XP tower,
Windows98 and below,
DOS,
8 bit guy,
Commodore 64,
Commodore- everything,
Z80,
IBM PC
and clones lover.

xouzouris

IGABMS wrote:
IGABMS wrote:
Why are we hosting this on another site?

You guys do realize we're like 100x more likely to accidentially DDOS our networks on accident.
*if we host it on our own devices.

Guys- has anybody else been having issues with getting notified of new posts?

Yes, scratch rarely notifies me of new posts, I have to check manually. I thought it was a feature, like when there are a lot of new posts in a forum it doesn't notify you because it might be annoying.

>> Head of Scratchedia <<

Scratchedia is the recreation of the
internet, on Scratch, that's safe!

Random Stuff:
My little cousin asked me why we can't make and run a nuclear reactor at home…

Don't forget to keep on Scratchin'

xouzouris

kkidslogin wrote:
A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).

First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.

>> Head of Scratchedia <<

Scratchedia is the recreation of the
internet, on Scratch, that's safe!

Random Stuff:
My little cousin asked me why we can't make and run a nuclear reactor at home…

Don't forget to keep on Scratchin'

kkidslogin

xouzouris wrote:
kkidslogin wrote:
A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.

Ha!

That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.

My siggy doesn't fit on ur screen! Select it and use CTRL (Or COMMAND if MAC) + SHIFT + DOWN to see the rest please.

when green flag clicked
forever
write code
repeat until <(bugs) = [0]>
test
fix bugs
end
release
end

Newest songs:
Hack Attack
Surrender
Deaf (NEW!)

My collaboration:
Starsoft Incorporated

What I am:
A proud BBcode user

Windows95 gamer on Windows XP tower,
Windows98 and below,
DOS,
8 bit guy,
Commodore 64,
Commodore- everything,
Z80,
IBM PC
and clones lover.

IGABMS

kkidslogin wrote:
xouzouris wrote:
kkidslogin wrote:
A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.

Hmmm. Interesting ideas!

However, I believe that Scratchedia should be de-centralized, hence not running on a rasberry pi.

I'm good for whatver.

IGABMS

kkidslogin wrote:
xouzouris wrote:
kkidslogin wrote:
A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.

This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

xouzouris

Ok, I have a Raspberry Pi if we need it.

>> Head of Scratchedia <<

Scratchedia is the recreation of the
internet, on Scratch, that's safe!

Random Stuff:
My little cousin asked me why we can't make and run a nuclear reactor at home…

Don't forget to keep on Scratchin'

kkidslogin

IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm………..

Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually

My siggy doesn't fit on ur screen! Select it and use CTRL (Or COMMAND if MAC) + SHIFT + DOWN to see the rest please.

when green flag clicked
forever
write code
repeat until <(bugs) = [0]>
test
fix bugs
end
release
end

Newest songs:
Hack Attack
Surrender
Deaf (NEW!)

My collaboration:
Starsoft Incorporated

What I am:
A proud BBcode user

Windows95 gamer on Windows XP tower,
Windows98 and below,
DOS,
8 bit guy,
Commodore 64,
Commodore- everything,
Z80,
IBM PC
and clones lover.

GAMS2

Don't tell me I have to scrap the website…

Edit: Also the problem with that is if the ISP (Internet Service Provider) went down, the “server” has a blackout, the internet box needs a reset, or the internet starts having problem on the server, its screwed. Hey, even if the “server” breaks, its all screwed.

Last edited by GAMS2 (May 17, 2021 15:44:09)

My signature is too big, select the text and use Ctrl + Shift + Down arrow! (I am the actual GAMS2! Gonna impersonate me for AFD? Don't include this line of text!)

Have I stopped recording?

No?
Then how do I turn this off?
Press the red button?
Okay I am pr-
Latest project:My Middle School Be Like: (WARNING: LOW QAULITY!)

Check out my new RP: Cops and Robbers RP! Hope to see yall there!

GAMS2 wrote:
If life throws something at you, try your best to dodge it, if it hits you and you fall…. get back up and keep moving forward. Don't give up, and make it through to the end. (MY OWN QUOTE!)

Do people still read these?

Nah, probably not.

IGABMS

kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm……….. Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually

Wow! THat's a lot of work to quote it manually!

I believe that if we want it to be accessible, people shouldn't need a rasberry pi. This should also be easily copy-able (with credit, of course)

It's up to people to store it. Whenever someone leaves, it just frees up more space for other people and lessens the load Scratch has to deal with. My vote is that if you leave, your website is cleared.

And guys-quick reminder: this is the INTERNET IN SCRATCH- NOT the internet partially rooted in Scratch but also runs on a Node.js server and requires each user to have a $15 rasberry pu.

IGABMS

IGABMS wrote:
kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm……….. Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually
Wow! THat's a lot of work to quote it manually!

I believe that if we want it to be accessible, people shouldn't need a rasberry pi. This should also be easily copy-able (with credit, of course)

It's up to people to store it. Whenever someone leaves, it just frees up more space for other people and lessens the load Scratch has to deal with. My vote is that if you leave, your website is cleared.

And guys-quick reminder: this is the INTERNET IN SCRATCH- NOT the internet partially rooted in Scratch but also runs on a Node.js server and requires each user to have a $15 rasberry pu.

You know what, why don't we just write this in Python and make this peer-to-peer with 256-bit asymmetric encryption and a remote data backup in 5 remote countries and have 3 more backups in a safety deposit box in a fortress?

This doesn't have to be so complex!

I have repeatedly stated that we DO NOT NEED to leave Scratch for this, yet it still seems that everyone insists that we do!

NILL_GAMES10

IGABMS wrote:
IGABMS wrote:
kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm……….. Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually
Wow! THat's a lot of work to quote it manually!

I believe that if we want it to be accessible, people shouldn't need a rasberry pi. This should also be easily copy-able (with credit, of course)

It's up to people to store it. Whenever someone leaves, it just frees up more space for other people and lessens the load Scratch has to deal with. My vote is that if you leave, your website is cleared.

And guys-quick reminder: this is the INTERNET IN SCRATCH- NOT the internet partially rooted in Scratch but also runs on a Node.js server and requires each user to have a $15 rasberry pu.

You know what, why don't we just write this in Python and make this peer-to-peer with 256-bit asymmetric encryption and a remote data backup in 5 remote countries and have 3 more backups in a safety deposit box in a fortress?

This doesn't have to be so complex!

I have repeatedly stated that we DO NOT NEED to leave Scratch for this, yet it still seems that everyone insists that we do!

Alright, lets have a poll tgen https://forms.gle/u4MeDkREDvx2AanE7

I've moved accounts. Go to my new account here.

Real_WeBino

Is this an OS? If not and it's a browser, then if you want it on an OS (which will hopefully one day be scripted on Visual Studio Code and added to HP, Dell, and chromebooks learn more here: Discover OS Forum)
I will be happy to add it

Hello! I am Real_WeBino! The owner of WeBino!

link images will be here soon when this account turns to scratcher

Check out my other profiles! GuitarGuyPlayz MrCreeperFX

Current possible DEV: WeBino Discover 1 - The OS Of The Future!
Next DEV: T.B.A
3rd DEV: T.B.A
4th DEV: T.B.A
5th DEV: T.B.A
6th DEV: T.B.A
7th DEV: T.B.A
8th DEV: T.B.A

kkidslogin

IGABMS wrote:
kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm……….. Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually
Wow! THat's a lot of work to quote it manually!

I believe that if we want it to be accessible, people shouldn't need a rasberry pi. This should also be easily copy-able (with credit, of course)

It's up to people to store it. Whenever someone leaves, it just frees up more space for other people and lessens the load Scratch has to deal with. My vote is that if you leave, your website is cleared.

And guys-quick reminder: this is the INTERNET IN SCRATCH- NOT the internet partially rooted in Scratch but also runs on a Node.js server and requires each user to have a $15 rasberry pu.

Hmmm… that makes something different than the internet… But it sounds cool. Like the internet before disk drives!

I have a question though: are you guys going to make a text version of this browser before graphical? A graphical browser can be quite difficult. And don't forget that the internet just looked like reading a file from UXTerm for a while.

My siggy doesn't fit on ur screen! Select it and use CTRL (Or COMMAND if MAC) + SHIFT + DOWN to see the rest please.

when green flag clicked
forever
write code
repeat until <(bugs) = [0]>
test
fix bugs
end
release
end

Newest songs:
Hack Attack
Surrender
Deaf (NEW!)

My collaboration:
Starsoft Incorporated

What I am:
A proud BBcode user

Windows95 gamer on Windows XP tower,
Windows98 and below,
DOS,
8 bit guy,
Commodore 64,
Commodore- everything,
Z80,
IBM PC
and clones lover.

NILL_GAMES10

kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm……….. Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually
Wow! THat's a lot of work to quote it manually!

I believe that if we want it to be accessible, people shouldn't need a rasberry pi. This should also be easily copy-able (with credit, of course)

It's up to people to store it. Whenever someone leaves, it just frees up more space for other people and lessens the load Scratch has to deal with. My vote is that if you leave, your website is cleared.

And guys-quick reminder: this is the INTERNET IN SCRATCH- NOT the internet partially rooted in Scratch but also runs on a Node.js server and requires each user to have a $15 rasberry pu.

Hmmm… that makes something different than the internet… But it sounds cool. Like the internet before disk drives! I have a question though: are you guys going to make a text version of this browser before graphical? A graphical browser can be quite difficult. And don't forget that the internet just looked like reading a file from UXTerm for a while.

I have sn idea…
It invilves using a sofisticated UI system and hiring a very good programmer

I've moved accounts. Go to my new account here.

xouzouris

IGABMS wrote:
IGABMS wrote:
kkidslogin wrote:
IGABMS wrote:
kkidslogin wrote:

xouzouris wrote:

kkidslogin wrote:

A more detailed explanation:

There are four cloud variables: server_data, server_response?, client_data, and client_response?. When the server sends data, it sets server_data to the data it is sending. It also sets server_response? to 1. Meanwhile, the client (the people that are online) is constantly scanning for when server_response? equals 1. When it becomes 1, the client stores the data somewhere and sets server_response? to 0 so that it doesn't read the data multiple times. When the client sends data, the same thing happens, except it uses client_data and client_response?, and the server sets client_response? to 0 after reading the data in client_data.

Presumably, the server uses a list to store data like websites. The downsides are that there can only be 200,000 websites (Unless you are running it in turbowarp), that every website is lost when the server loses power, and that there is a limited amount of memory availible for websites (~1GB if using a Raspberry Pi 4 2GB).
First of all, I THOUGHT OF THE EXACT SAME THING O_O Second, I do have a raspberry pi.
Ha! That's funny! I also have an add-on to this idea: if you don't want Chromium running 24/7, you can package a standalone offline package using Turbowarp Packager, and have cloud variables set to “Turbowarp's Online Server”. And perhaps pester
-GarboMuffin- about adding cloud variables and locally stored variables to TurboWarp Desktop.
This is what I've been saying, except without another central computer.

This could be accomplished by having a data transfer variable and a request address
and the client is always checking for a request to its address and then sets the data transfer variable to the content.

Hmmm……….. Having the client host the website could be sketchy… When the client leaves the project or even his computer falls asleep, the website would be lost forever. So really, the client would need a raspberry pi running server mode as well, and preferably a separate account to run the server.

I forgot to quote your post so I had to do it manually
Wow! THat's a lot of work to quote it manually!

I believe that if we want it to be accessible, people shouldn't need a rasberry pi. This should also be easily copy-able (with credit, of course)

It's up to people to store it. Whenever someone leaves, it just frees up more space for other people and lessens the load Scratch has to deal with. My vote is that if you leave, your website is cleared.

And guys-quick reminder: this is the INTERNET IN SCRATCH- NOT the internet partially rooted in Scratch but also runs on a Node.js server and requires each user to have a $15 rasberry pu.

You know what, why don't we just write this in Python and make this peer-to-peer with 256-bit asymmetric encryption and a remote data backup in 5 remote countries and have 3 more backups in a safety deposit box in a fortress?

This doesn't have to be so complex!

I have repeatedly stated that we DO NOT NEED to leave Scratch for this, yet it still seems that everyone insists that we do!

The scratch cloud variable servers are just not good enough! They are slow and unreliable!

Last edited by xouzouris (May 17, 2021 19:28:13)

>> Head of Scratchedia <<

Scratchedia is the recreation of the
internet, on Scratch, that's safe!

Random Stuff:
My little cousin asked me why we can't make and run a nuclear reactor at home…

Don't forget to keep on Scratchin'

Discuss Scratch