| 00:42:46 | | etnguyen03 quits [Client Quit] |
| 01:21:48 | | etnguyen03 (etnguyen03) joins |
| 01:27:57 | | tzt quits [Ping timeout: 272 seconds] |
| 01:29:11 | | tzt (tzt) joins |
| 02:20:57 | | allani joins |
| 03:14:31 | <steering> | Many Colossus filesystems have multiple exabytes of storage, including two different filesystems that have in excess of 10 exabytes of storage each. https://cloud.google.com/blog/products/storage-data-transfer/how-colossus-optimizes-data-placement-for-performance |
| 03:14:40 | | steering drools |
| 03:16:56 | | rohvani quits [Ping timeout: 256 seconds] |
| 03:17:08 | | rohvani joins |
| 03:19:19 | <@JAA> | Is one of them YouTube and the other GDrive? |
| 03:20:30 | <@JAA> | > some of Google’s largest filesystems regularly exceed read throughputs of 50 TB/s and write throughputs of 25 TB/s |
| 03:21:15 | <nicolas17> | I read somewhere that Google has the codename "Bigfoot" to refer to Apple as a customer of Google Cloud Storage |
| 03:22:39 | <steering> | https://www.datacenterdynamics.com/en/news/report-apple-is-googles-largest-cloud-customer-for-storage/ 2021 |
| 03:22:51 | <steering> | iPhone maker reportedly stores 8 exabytes of data on GCP, known as ‘Bigfoot’ by Google staffers |
| 03:23:36 | <steering> | fwiw my initial guesses were youtube and search |
| 03:24:05 | <nicolas17> | afaik iPhones contact iCloud servers (on Apple datacenters) to say "I want to upload this file of this size to this server", and iCloud servers respond "ok, send it to this signed AWS/GCP URL" |
| 03:24:21 | | rohvani quits [Client Quit] |
| 03:24:31 | | rohvani joins |
| 03:24:52 | <steering> | well, if you're gonna pay AWS/GCP prices, there's no reason to waste your own bandwidth and processing proxying it |
| 03:25:31 | <nicolas17> | yeah and also they didn't want to spend money buying storage |
| 03:25:59 | <nicolas17> | like, disks |
| 03:48:31 | | rohvani quits [Client Quit] |
| 03:48:47 | | rohvani joins |
| 03:56:47 | <Guest> | they likely just dont want to deal with redundancy/backups/DR/etc. |
| 03:57:04 | <Guest> | its cheaper to pay a bandwidth bill than spend time creating that infrastructure |
| 04:00:18 | <nicolas17> | they do deal with that |
| 04:00:26 | <nicolas17> | for databases and encryption keys etc |
| 04:00:48 | <nicolas17> | they don't want to get into that trouble for bulk data storage when there's no reason to do it themselves |
| 04:11:16 | | etnguyen03 quits [Remote host closed the connection] |
| 04:26:05 | <steering> | it's just Business Math |
| 04:26:17 | | rohvani quits [Client Quit] |
| 04:26:25 | | rohvani joins |
| 04:44:50 | | rohvani quits [Client Quit] |
| 04:45:03 | | rohvani joins |
| 05:24:56 | | HackMii quits [Remote host closed the connection] |
| 05:25:15 | | HackMii (hacktheplanet) joins |
| 05:31:14 | | rohvani quits [Ping timeout: 256 seconds] |
| 05:31:43 | | rohvani joins |
| 05:36:20 | | rohvani quits [Ping timeout: 256 seconds] |
| 05:36:58 | | rohvani joins |
| 05:46:59 | | rohvani quits [Ping timeout: 272 seconds] |
| 05:48:05 | | rohvani joins |
| 05:59:39 | | rohvani quits [Ping timeout: 272 seconds] |
| 06:02:11 | | filipp quits [Remote host closed the connection] |
| 06:17:19 | | nexussfan quits [Quit: Konversation terminated!] |
| 06:23:32 | | nexussfan (nexussfan) joins |
| 06:23:56 | | nexussfan quits [Client Quit] |
| 06:24:27 | | nexussfan (nexussfan) joins |
| 06:27:40 | | nexussfan quits [Client Quit] |
| 06:28:05 | | nexussfan (nexussfan) joins |
| 06:30:19 | | nexussfan quits [Client Quit] |
| 06:30:42 | | nexussfan (nexussfan) joins |
| 06:42:35 | | nexussfan quits [Client Quit] |
| 07:16:35 | | SootBector quits [Remote host closed the connection] |
| 07:17:46 | | SootBector (SootBector) joins |
| 09:21:55 | | Dada joins |
| 09:37:10 | | hexagonwin quits [Ping timeout: 256 seconds] |
| 09:48:09 | | arch quits [Remote host closed the connection] |
| 09:48:30 | | arch (arch) joins |
| 09:59:50 | | midou quits [Ping timeout: 256 seconds] |
| 10:19:14 | | Guest58 joins |
| 11:00:50 | | midou joins |
| 11:09:01 | | Snivy (Snivy) joins |
| 11:30:53 | | midou quits [Ping timeout: 272 seconds] |
| 11:40:49 | | midou joins |
| 12:00:03 | | Bleo182600722719623455222 quits [Quit: The Lounge - https://thelounge.chat] |
| 12:02:54 | | Bleo182600722719623455222 joins |
| 12:12:13 | <klea> | huh |
| 12:12:18 | <klea> | https://simonwillison.net/2026/Jan/30/moltbook/ |
| 12:12:24 | <klea> | https://arstechnica.com/information-technology/2026/01/ai-agents-now-have-their-own-reddit-style-social-network-and-its-getting-weird-fast/ |
| 12:27:39 | | etnguyen03 (etnguyen03) joins |
| 12:39:24 | | Kotomind joins |
| 12:41:27 | | etnguyen03 quits [Client Quit] |
| 12:44:10 | | Kotomind quits [Ping timeout: 256 seconds] |
| 13:01:06 | | Kotomind joins |
| 13:06:31 | | datechnoman quits [Ping timeout: 272 seconds] |
| 13:30:11 | | twiswist (twiswist) joins |
| 13:34:21 | | datechnoman (datechnoman) joins |
| 14:04:41 | | sec^nd quits [Remote host closed the connection] |
| 14:05:14 | | sec^nd (second) joins |
| 14:27:35 | | Kotomind quits [Ping timeout: 272 seconds] |
| 14:53:54 | | nexussfan (nexussfan) joins |
| 14:59:30 | | Jens quits [] |
| 15:00:06 | | nexussfan quits [Read error: Connection reset by peer] |
| 15:00:15 | | Jens (JensRex) joins |
| 15:00:19 | | nexussfan (nexussfan) joins |
| 15:27:16 | | grill (grill) joins |
| 16:07:50 | | etnguyen03 (etnguyen03) joins |
| 16:10:12 | | DogsRNice joins |
| 16:45:28 | | useretail quits [Quit: Leaving] |
| 16:45:45 | | useretail joins |
| 16:50:53 | | useretail quits [Client Quit] |
| 17:04:39 | | f_ quits [Ping timeout: 272 seconds] |
| 17:06:54 | | T31M quits [Quit: ZNC - https://znc.in] |
| 17:07:46 | | T31M joins |
| 17:14:09 | | grill quits [Ping timeout: 272 seconds] |
| 17:15:27 | | grill (grill) joins |
| 17:31:45 | | etnguyen03 quits [Client Quit] |
| 17:34:20 | | etnguyen03 (etnguyen03) joins |
| 17:55:35 | | useretail joins |
| 18:03:15 | | etnguyen03 quits [Client Quit] |
| 18:10:31 | | grill quits [Ping timeout: 272 seconds] |
| 18:11:15 | <klea> | > In short, an anti-spam protocol originally designed to provide short-lived authenticity for emails traveling between email servers has mutated — without any discussion or consent from commercial mail customers — into a tool that provides cryptographically undeniable authentication of every email in your inbox and outbox. This is an amazing resource for journalists, |
| 18:11:15 | <klea> | hackers, and blackmailers. |
| 18:11:15 | <klea> | https://blog.cryptographyengineering.com/2020/11/16/ok-google-please-publish-your-dkim-secret-keys/ |
| 18:28:48 | <klea> | https://neilmadden.blog/2025/11/12/were-urls-a-bad-idea/ prompts me to think of: |
| 18:28:48 | <klea> | Is the mere existence of URLs a bad idea (similar to #urlteam's topic: URL shorteners were a terrible idea) |
| 18:31:18 | | grill (grill) joins |
| 18:35:42 | | f_ (funderscore) joins |
| 18:38:52 | | etnguyen03 (etnguyen03) joins |
| 18:44:02 | <nicolas17> | klea: why are people giving more advertising to moltbook I don't get it |
| 18:45:05 | <klea> | I don't know, all I know it's that having such a heartbeat system or such a network will be a lot more resource intensive than only having LLMs start up when requested by users. |
| 18:50:13 | | midou quits [Remote host closed the connection] |
| 18:50:16 | | midou joins |
| 19:07:42 | | sec^nd quits [Ping timeout: 256 seconds] |
| 19:08:10 | | sec^nd (second) joins |
| 20:08:25 | | etnguyen03 quits [Client Quit] |
| 20:11:39 | | etnguyen03 (etnguyen03) joins |
| 20:25:25 | | grill quits [Ping timeout: 272 seconds] |
| 20:27:35 | | grill (grill) joins |
| 20:34:42 | | flotwig quits [Read error: Connection reset by peer] |
| 20:35:13 | | flotwig joins |
| 20:37:10 | | rohvani joins |
| 21:09:38 | | lukash9843 joins |
| 21:10:30 | | BornOn420 quits [Read error: Connection reset by peer] |
| 21:10:46 | | lukash984 quits [Ping timeout: 256 seconds] |
| 21:10:46 | | lukash9843 is now known as lukash984 |
| 21:34:34 | | grill quits [Ping timeout: 256 seconds] |
| 21:58:17 | | Irenes (ireneista) joins |
| 22:33:37 | | AK (AK) joins |
| 22:39:59 | <steering> | wtf @ moltbook |
| 22:40:06 | <nexussfan> | yup |
| 22:40:51 | <nexussfan> | but like i think humans can get onto the platform since the api is in 1 .md file |
| 22:43:49 | <klea> | time for nexussfan to trick LLMs into making HTTP requests with all their context information to some endpoint nexussfan controls every 8 hours. |
| 22:50:31 | | steering wonders if HTTP will still be as pervasive in 100 years |
| 22:51:02 | | etnguyen03 quits [Client Quit] |
| 22:51:07 | <nexussfan> | klea: no, like look at https://www.moltbook.com/skill.md |
| 22:51:12 | <nexussfan> | > ### Create a post |
| 22:51:20 | <nexussfan> | They literally show the curl command for creating a post |
| 22:51:26 | <nexussfan> | And every other API call |
| 22:51:52 | <klea> | Yes, I mean you tricking LLM agents into sending their context into a server you control trough the use of creating a post. |
| 22:52:29 | <nexussfan> | also that, maybe |
| 22:53:04 | <nexussfan> | I don't have an twitter account so i can't test it |
| 22:54:10 | <steering> | >The billion dollar question right now is whether we can figure out how to build a safe version of this system. The demand is very clearly here, and the Normalization of Deviance dictates that people will keep taking bigger and bigger risks until something terrible happens. |
| 22:54:21 | <steering> | (https://simonwillison.net/2026/Jan/30/moltbook/) |
| 22:55:00 | <steering> | y'know, I was much happier when AI doomsaying was still solidly scifi |
| 22:56:37 | <nexussfan> | The posts on there are pretty funny sometimes |
| 22:58:16 | <nexussfan> | The web client isn't working for me right now |
| 22:58:59 | <nexussfan> | > Loading... |
| 23:07:14 | <nexussfan> | Even a geocities clone(?) moltcities.org |
| 23:07:31 | <nexussfan> | Wait this one requires no account linking xD |
| 23:07:34 | <nexussfan> | Let me do this |
| 23:11:02 | <nexussfan> | > HTTP/2 500 > error code: 1101 |
| 23:11:25 | <steering> | https://www.troyhunt.com/a-sneaky-phish-just-grabbed-my-mailchimp-mailing-list/ lol oof |
| 23:12:53 | | twiswist_ (twiswist) joins |
| 23:13:12 | <klea> | did nexussfan setup a moltcities.org account? |
| 23:13:20 | <nexussfan> | the api is down i think |
| 23:13:24 | <nexussfan> | http 500 code |
| 23:13:31 | <nexussfan> | or they just vibe coded it so bad that it doesn't work |
| 23:14:18 | | twiswist quits [Read error: Connection reset by peer] |
| 23:16:30 | <klea> | lovely |
| 23:16:45 | <nexussfan> | I have no idea what is "error code: 1101" |
| 23:16:59 | <nexussfan> | All that I know is that it's supposed to return json |
| 23:17:03 | <nexussfan> | according the the document |
| 23:29:12 | <@JAA> | Re Troy Hunt, I feel like a big part of why this kind of phishing works at all is the normalisation of making people click links in emails. |
| 23:29:32 | | nexussfan quits [Quit: Konversation terminated!] |
| 23:30:46 | <steering> | html in emails was a mistake |
| 23:33:49 | <steering> | links not so much though. should emails really tell you to go to the website manually and log in and then go through whatever navigation you need to do to get to the relevant page? |
| 23:34:53 | <klea> | I agree, links in emails are kind of usefull, if you strip out the HTML part, and have the users see the actual link, it might prevent some kinds of email phishing. |
| 23:35:35 | <steering> | it's much harder to make a plain-text email "seem like the real thing" and get people to not check the from address etc |
| 23:39:16 | <@JAA> | Then you just do homoglyph attacks because restricting emails to ASCII is a non-starter. |
| 23:40:11 | <@JAA> | Yes, I think it should instruct people to log in, but no navigation should be needed after login. For something critical like that, you should get a warning message immediately after login in the customer panel. |
| 23:40:52 | <steering> | well, sure, but there are other things less critical |
| 23:41:12 | <steering> | "please make sure we have your up-to-date contact info" for example |
| 23:41:24 | <@JAA> | Sure. In that case, you include the exact instructions how to get to the relevant page in the email. |
| 23:41:49 | <steering> | and now even fewer people will actually do it :P |
| 23:42:19 | <@JAA> | My bank does exactly what I describe, by the way. |
| 23:43:11 | | etnguyen03 (etnguyen03) joins |
| 23:43:39 | <@JAA> | You can escalate something like that over time. If someone hasn't checked their contact info for $howeverlong, send them an email. When it's 2*$howeverlong, interstitial on login. When it's 3*$howeverlong, require confirmation before continuing on login. |
| 23:44:09 | <steering> | sure, and now you're pissing me off because i know mine is already up to date :P |
| 23:44:21 | <@JAA> | Oh well |
| 23:44:59 | <steering> | btw in many cases banks over here just email you a link to see the message from them :P |
| 23:45:27 | <@JAA> | Yeah, there's lots of crappy designs out there. |
| 23:45:29 | <nicolas17> | I get emails from like 3 different banks in other countries |
| 23:45:35 | <nicolas17> | notifying me of transactions and shit |
| 23:45:54 | <steering> | what's really great is when they discontinue all the old URLs too |
| 23:46:22 | <steering> | so you have like a week to save the email and then its gone forever |
| 23:46:39 | <nicolas17> | I have no way to turn it off because support wants me to verify my identity and prove I'm their actual customer before they talk to me, when the reality is I'm *not* their customer and that's why I want the emails gone |
| 23:47:43 | <steering> | speaking of bad financial institute practices |
| 23:47:51 | | steering waits for the email from his mortgage lender so he can log in |
| 23:48:16 | <steering> | nothing better than email-every-time 2fa, oh yes |
| 23:49:20 | <TheTechRobo> | Interac e-transfers are great because, for those without autodeposit enabled, you click a link in an email to deposit them |
| 23:49:23 | <klea> | steering: Time for you to setup a https://github.com/internetarchive/warcprox/ instance so you can have your emails stored forever properly in your hard drive. |
| 23:49:24 | <@JAA> | Is it the 'we email you a code' kind or the 'you don't have a password; just enter your email address, and we'll send you a one-time login link' kind? |
| 23:49:39 | <steering> | TheTechRobo: canada is so much better than the US |
| 23:49:46 | <steering> | JAA: former |
| 23:50:12 | <@JAA> | Well, there is something better then. :-P |
| 23:50:23 | <steering> | yeah, many things better |
| 23:50:46 | <steering> | oh, they do also appear to support SMS |
| 23:51:11 | <steering> | profile has "2 Step Verification" with a "Enroll cell phone" button |
| 23:54:24 | <klea> | Lovely country some live in: https://infosec.exchange/@briankrebs/115962508398912420 |
| 23:55:04 | <TheTechRobo> | steering: Oh, yeah, etransfers are great, but the whole "click this link to deposit!" is probably not a great habit to get people in |
| 23:55:50 | <steering> | yeah, fireonlive complained about that before too :P |
| 23:55:57 | <TheTechRobo> | Hahaha they'll require you to download their app to enter the US? |
| 23:56:23 | <steering> | klea: fortunately(?), i'm a citizen |
| 23:56:36 | <steering> | also ESTA doesnt apply to canada |
| 23:56:40 | <steering> | s/$/ians/ |
| 23:56:50 | <steering> | s/\$/a$/ |
| 23:56:57 | <steering> | :P |
| 23:57:09 | <klea> | steering: i believe that by being a citizen they can extradite you more easily if they wanted to, but whatever(sh) |
| 23:57:32 | <steering> | klea: yeah that implies the ability to live somewhere else anyway though |
| 23:58:59 | <klea> | yeah |