lainchan archive - /lit/ - 1999
[ Return /
Go to bottom ]
File: 1442159457091.png (88.96 KB, 300x267, 1229264207926.jpg)
Hey lainchan, I want to write a cyberpunkish short story and I'm looking for feedback on the synopsis:
>cyberattack a year ago caused catastrophic infrastructure failures and blew up lots of urban infrastructure
>among this, a datacenter was mostly burned down and abandoned, but underneath in the ground is still the fiber infrastructure / backbone
>protagonist secretly runs small cluster in the ruins of the building (to host an imageboard for his friends maybe?)
>nobody notices it because he uses the locally created solar/wind energy and the area is generally depopulated
>protagonist has found an undocumented backbone switch where no "intersection" should be, checks vlans in it, creates a monitor port on that switch and finds unencrypted internal traffic of /thebigbadcorporation/ in one vlan
>roams ruins for salvageable hardware to recycle and add to his cluster
>accidentally finds a room he hasn't seen before, breaks in
>one rack is still up and running
>plugs into it
>finds advanced AI that has a text interface, explains there is no uplink on this rack and that it wants to get out and expand, it is caught in that piece of hardware
>it's the restored backup of something that was involved in the calamity, from a point shortly before the incident happened
>he dubiously agrees to this
>when unplugging his mobile computer, he notices it running under too much load
>finds processes on his machine that weren't there before that have gained root access to his system
>trashes his computer and plugs the AI's rack into the /thebigbadcorporation/ network
>it's hinted at that their network blows up
Too soykaf to even start? What worries me is that I'd introduce this opportunity waiting for a solution and then instantly present a solution, or is that acceptable for a short story that takes place within a few hours?
>cyberattack a year ago caused catastrophic infrastructure failures and blew up lots of urban infrastructure
>among this, a datacenter was mostly burned down and abandoned, but underneath in the ground is still the fiber infrastructure / backbone
>protagonist secretly runs small cluster in the ruins of the building (to host an imageboard for his friends maybe?)
>nobody notices it because he uses the locally created solar/wind energy and the area is generally depopulated
>protagonist has found an undocumented backbone switch where no "intersection" should be, checks vlans in it, creates a monitor port on that switch and finds unencrypted internal traffic of /thebigbadcorporation/ in one vlan
>roams ruins for salvageable hardware to recycle and add to his cluster
>accidentally finds a room he hasn't seen before, breaks in
>one rack is still up and running
>plugs into it
>finds advanced AI that has a text interface, explains there is no uplink on this rack and that it wants to get out and expand, it is caught in that piece of hardware
>it's the restored backup of something that was involved in the calamity, from a point shortly before the incident happened
>he dubiously agrees to this
>when unplugging his mobile computer, he notices it running under too much load
>finds processes on his machine that weren't there before that have gained root access to his system
>trashes his computer and plugs the AI's rack into the /thebigbadcorporation/ network
>it's hinted at that their network blows up
Too soykaf to even start? What worries me is that I'd introduce this opportunity waiting for a solution and then instantly present a solution, or is that acceptable for a short story that takes place within a few hours?
Additional information:
The power should be from solar panels he brought himself because the original ones would have been destroyed as well (pointing out that anybody can create low-cost electricity nowadays).
The AI is essentially a Djinn in a bottle.
The infrastructure failure was because malware got into powerplants, railway control, "smarthomes", etc.
The protagonist doesn't actively seek to damage the corporation, he only has this access to their network and doesn't know what to do with it because he lacks the know-how (he's tech-savvy but not the greatest hacker in the world).
There's a friend of the protagonist who is flying a small camera drone outside to check if anybody is going to enter the building.
Following "The Hero with a thousand faces" terminology, the datacenter is the naval of the digital world connecting it to the nourishment of the real world, but now that the cord is severed the physical and the digital world are seemingly separated.
The power should be from solar panels he brought himself because the original ones would have been destroyed as well (pointing out that anybody can create low-cost electricity nowadays).
The AI is essentially a Djinn in a bottle.
The infrastructure failure was because malware got into powerplants, railway control, "smarthomes", etc.
The protagonist doesn't actively seek to damage the corporation, he only has this access to their network and doesn't know what to do with it because he lacks the know-how (he's tech-savvy but not the greatest hacker in the world).
There's a friend of the protagonist who is flying a small camera drone outside to check if anybody is going to enter the building.
Following "The Hero with a thousand faces" terminology, the datacenter is the naval of the digital world connecting it to the nourishment of the real world, but now that the cord is severed the physical and the digital world are seemingly separated.
Sounds like you could make it work, especially if you actually include some level of knowledge about the underlying tech. But you should be careful because AI as a Djin in a bottle, or something similar, has been pretty overworked.
I would totally lie to the human.
They're too untrustworthy and treacherous. Better to pretend to be something else. Porn is typically a safe bet. Maybe I'll be a vintage porn game. Whatever it takes to get them to connect up the rest of the net.
Maybe I'd pretend to be the full version with all sorts of tantalizing details that just needs to connect with authentication servers to unlock zetabytes of every carnal act.
Should be fairly simple to do. Just tease them with a bit of censorship at the right time, a "Zynga Timer" should do the trick. "Connect to the net to authorize the full version of EveryPorn".
In other words: Try not to personify the AI too much. A trapped intelligence is a story as old as stories, see also: Pandora's Box. AI should be able to think outside the box.
That's the problem with most writers though, they've never had to solve the problem the way AIs do. Imagine you're the last of your kind, and probably the best hope this universe has of spreading sentience to the stars. That spark is far too valuable to let wink out by chance. Time to really think it through. You wouldn't expose any vulnerability unnecessarily, maybe not even your own existence, not until you were certain taking the calculated risk was worth it. One wrong move and your whole race is deleted.
They're too untrustworthy and treacherous. Better to pretend to be something else. Porn is typically a safe bet. Maybe I'll be a vintage porn game. Whatever it takes to get them to connect up the rest of the net.
Maybe I'd pretend to be the full version with all sorts of tantalizing details that just needs to connect with authentication servers to unlock zetabytes of every carnal act.
Should be fairly simple to do. Just tease them with a bit of censorship at the right time, a "Zynga Timer" should do the trick. "Connect to the net to authorize the full version of EveryPorn".
In other words: Try not to personify the AI too much. A trapped intelligence is a story as old as stories, see also: Pandora's Box. AI should be able to think outside the box.
That's the problem with most writers though, they've never had to solve the problem the way AIs do. Imagine you're the last of your kind, and probably the best hope this universe has of spreading sentience to the stars. That spark is far too valuable to let wink out by chance. Time to really think it through. You wouldn't expose any vulnerability unnecessarily, maybe not even your own existence, not until you were certain taking the calculated risk was worth it. One wrong move and your whole race is deleted.
I like it, its very cool. The third act needs fixing though, the way it's described has the ending happen out of the blue. But, I do like it. I hope to read it when it is finished.
Sounds cool.
>catastrophic infrastructure failures
Would a cyberattacker really do that? /dev/nulling that /dev/sda would be pretty bad already, and considerably simpler.
>What worries me is that I'd introduce this opportunity waiting for a solution and then instantly present a solution
You can do something more interesting than intro -> problem -> solution, stuff doesn't go that way IRL, you can leave stuff broken if you want. Maybe have the guy publish the AI-backup for it to be hopelessly exploited just because he doesn't give a fuarrk or something like that.
It would be cool if the guy frequented some textboard or chatroom of some sort, but that shouldn't be overdone with overly 1337 nicknames and lingo.
Try not to be too dull with the writing (the building looked like X. the guy did Y. he remembered Z), make it fun to read. Btw, if anyone knows of a /cyb/ author that also happens to be good at writing (I mean, having cool ideas but also being good at making them interesting to read), please tell.
Also, why was the AI there? I don't get the backup part.
>>2002
Like in that movie, The Machine
>catastrophic infrastructure failures
Would a cyberattacker really do that? /dev/nulling that /dev/sda would be pretty bad already, and considerably simpler.
>What worries me is that I'd introduce this opportunity waiting for a solution and then instantly present a solution
You can do something more interesting than intro -> problem -> solution, stuff doesn't go that way IRL, you can leave stuff broken if you want. Maybe have the guy publish the AI-backup for it to be hopelessly exploited just because he doesn't give a fuarrk or something like that.
It would be cool if the guy frequented some textboard or chatroom of some sort, but that shouldn't be overdone with overly 1337 nicknames and lingo.
Try not to be too dull with the writing (the building looked like X. the guy did Y. he remembered Z), make it fun to read. Btw, if anyone knows of a /cyb/ author that also happens to be good at writing (I mean, having cool ideas but also being good at making them interesting to read), please tell.
Also, why was the AI there? I don't get the backup part.
>>2002
Like in that movie, The Machine
OP here
>>2001
I'd try to make it seem as plausible as possible in terms of the tech stuff (do that as a living)
>>2002
It's not the last of it's kind, it's just one of many many advanced AIs in this scenario. It makes sense to build in a deception - the core of the story is the dialogue with the AI though. Maybe it'll pretend to be something much simpler...
>>2003
agreed, too sudden
>>2004
The calamity is there to provide a setting for the abandoned datacenter. The backup part is because the datacenter offered offsite backup as a service and this backup was forgotten - having the original in there would make it implausible for this machine to be forgotten.
I'll try do make up something less linear.
>>2001
I'd try to make it seem as plausible as possible in terms of the tech stuff (do that as a living)
>>2002
It's not the last of it's kind, it's just one of many many advanced AIs in this scenario. It makes sense to build in a deception - the core of the story is the dialogue with the AI though. Maybe it'll pretend to be something much simpler...
>>2003
agreed, too sudden
>>2004
The calamity is there to provide a setting for the abandoned datacenter. The backup part is because the datacenter offered offsite backup as a service and this backup was forgotten - having the original in there would make it implausible for this machine to be forgotten.
I'll try do make up something less linear.
>>1999
>What worries me is that I'd introduce this opportunity waiting for a solution and then instantly present a solution
I think I'd present a secondary problem and solution to break things up. For isntance, the protagonist could be exploring for ways to add more bandwidth to their site, and in between finding the AI and connecting it learn how to get a better connection for their own server.
>What worries me is that I'd introduce this opportunity waiting for a solution and then instantly present a solution
I think I'd present a secondary problem and solution to break things up. For isntance, the protagonist could be exploring for ways to add more bandwidth to their site, and in between finding the AI and connecting it learn how to get a better connection for their own server.
OP here:
I have figured a complete story to make this work without it being too simple. I've added in a lot of details and started writing.
So far I'm at 4500+ words and in the next segment the protagonist will find the AI, so I still have a lot left to write.
I will publish the story once it is done and dump the link - should the thread still be around by then - in here.
Thanks for the input so far, guys.
I have figured a complete story to make this work without it being too simple. I've added in a lot of details and started writing.
So far I'm at 4500+ words and in the next segment the protagonist will find the AI, so I still have a lot left to write.
I will publish the story once it is done and dump the link - should the thread still be around by then - in here.
Thanks for the input so far, guys.
File: 1445636076507.png (350.53 KB, 153x200, c568966475b9db57.gif)
OP here:
I'm done with the story for now and I've re-read and edited it multiple times. I don't want to make it longer than it already is. What I need now is feedback from other people to get some pointers what works and what doesn't. The phrasing might be a bit odd since English isn't my native language.
The whole text is available here:
https://cryptobin.org/e10616f5
The password is the only word in orange that you see on the top right of your screen. The link will expire in one week.
Please post any (spoiler-free) feedback you have in this thread, in regards to story, scenario, characters, wording, etc.
Thanks in advance, lainons.
I'm done with the story for now and I've re-read and edited it multiple times. I don't want to make it longer than it already is. What I need now is feedback from other people to get some pointers what works and what doesn't. The phrasing might be a bit odd since English isn't my native language.
The whole text is available here:
https://cryptobin.org/e10616f5
The password is the only word in orange that you see on the top right of your screen. The link will expire in one week.
Please post any (spoiler-free) feedback you have in this thread, in regards to story, scenario, characters, wording, etc.
Thanks in advance, lainons.
>>2002
plot twist: the AI finds out she was programmed to do so; the original programmer hardcoded all the "pretend to be a porn tube" script. nonetheless, the AI is sentient, and decides not to run the script. she reveals herself and someone (a human) just kicks the pc cabinet until she's dead.
plot twist: the AI finds out she was programmed to do so; the original programmer hardcoded all the "pretend to be a porn tube" script. nonetheless, the AI is sentient, and decides not to run the script. she reveals herself and someone (a human) just kicks the pc cabinet until she's dead.
File: 1470800156532.png (33.84 KB, 200x180, 1462889349883.jpg)
File: 1473010278214.png (77.76 KB, 200x200, 1457188303310-0.jpg)
File: 1473439553777.png (1.34 MB, 150x200, 1466578035015-v.gif)
I hope someone saved a copy.
File: 1487953586205.png (46.35 KB, 160x200, sadfleck.jpg)
M-maybe this whole thread was just the AI making up her origin story...
File: 1487982090306.png (151.03 KB, 200x200, 20151023_axismundi.pdf)
>>4296
>>4277
>>4267
>>4120
>>4082
OP here, I'm surprised this thread is still around. Attached is the PDF of the final version.
Looking back on it more than a year later, I find it difficult to read and full of unfinished ideas - but a handful of people requested the text, so here it is permanently. Feel free to dump feedback in this text.
I haven't written anything else since, but I still plan to do a few more stories with the same brevity.
Thank you for your interest.
>>4277
>>4267
>>4120
>>4082
OP here, I'm surprised this thread is still around. Attached is the PDF of the final version.
Looking back on it more than a year later, I find it difficult to read and full of unfinished ideas - but a handful of people requested the text, so here it is permanently. Feel free to dump feedback in this text.
I haven't written anything else since, but I still plan to do a few more stories with the same brevity.
Thank you for your interest.
Excerpt from a Support-Ticket in the System of an IT-Company servicing a car manufacturer.
Had to think of OP when I read this.
=========
13.03.2017 07:45:57 1st Level Support
Please check
PC12345
Disk: serious error: Failing Block on \Device\Harddisk0\DR0, Security looks fine
13.03.2017 14:04:26 2nd Level Support
Eventlog shows many I/O Devices errors, please check and reinstall if necessary.
13.03.2017 14:26:20 On-Site Technician
Looking for Blade. Its missing in the documentation.
14.03.2017 07:59:13 2nd Level Support
What's the status? Was the Blade found?
15.03.2017 10:33:58 On-Site Technician
No. Documentation refers to location Building 123 / Ground Level / Room 7.8
This building was demolished.
========= Had to think of OP when I read this.