#indieweb 2025-01-29
2025-01-29 UTC
alfiee and jbrr joined the channel
# 
bossy Thoughts on getting a short unique domain that has cheap renewals for email? I’ve had a micro.domains
jbrr joined the channel
# bossy *had a look at
alfiee, alephalpha0, tavare and grufwub joined the channel
# 
[aciccarello] I've been thinking about a short domain too
# [aciccarello] I bought 2 already but wasn't happy with how easy they are to say out loud. Which is important if I want to use it for an email address too.
tavare, lockywolf, alfiee, helveticamono, cedric, antranigv_, btrem, jbrr, mathilde, jumper and cedric_ joined the channel
jbrr joined the channel
alfiee joined the channel
# mathilde is someone looking for a domain name? have you heard of https://nic.eu.org?
# bossy Some email hosters don’t like you using free domains
# vinceaggrippino mathilde++ Thank you. I almost wasn't able to build an IndieWeb compatible site because I can't afford a domain. It turns out that I have some time left on the domain I registered when I still had an income. So it worked out for me but there may be some others like me who can't handle _any_ expenses right now.
alfiee joined the channel
[Murray] joined the channel
# 
[Murray] woah how expensive is your phone service 😬 I could run my phone for almost 4 months for the annual cost of a .http://co.uk domain, which I feel are pretty cheap (maybe that's no longer the case?)
# vinceaggrippino It's a perfectly reasonable cost. I'm just in sort of an unfortunate situation. I can't afford a mobile phone, either. My phone service (approximately $23/mo) is _subsidized_ by a family member who may need me to be available in case of emergency, but we've had to discuss cutting the phone service and relying on WhatsApp with WiFi more than once.
# vinceaggrippino Oh, and regarding coffee drinks, I never partake of the starbucksian variety. I'm very happy with my Nescafe Gold, but that doesn't stop me from being a little jealous when I see [tantek] use his espresso machine during the HWC video calls 😅
# vinceaggrippino hehe I'm in Malaysia. My phone bill is actually 90 ringgit per month.
alfiee joined the channel
# bossy Just registered my domain. Got a humdinger. Anyone who knows me or has met me would be able to get my name from it, but for anyone else it’s cryptic enough to have some anonymity
# bossy rss.gdn
# bossy You can prob guess my first name from that
jbrr, alfiee, ttybitnik and tyil joined the channel
# mathilde [tantek]: thanks capitalism, here in france we got unlimited calls, 5G bandwidth from and to the whole european union for 20€/month.
hacknorris joined the channel
# hacknorris just asking. should we all make blogs if we know it will be scraped by ais?
# mathilde hacknorris: I know it doesn't stop them all, but I'm banning all the OpenAI IP addresses pool on the firewall level, and serving a very complete robots.txt file against them.
# hacknorris what about github/codeberg pages people?
# hacknorris no server, just html files. no ability to filter anything from backends
# hacknorris (or neocities)
# hacknorris mathilde?
# hacknorris if only plain robots.txt can be done? none of ais read it anyway…
# mathilde hacknorris: then you have to rely to robots.txt only and pray for them to honor your requests LoL
# hacknorris that's why i ask if people should have blogs in these times…
# mathilde but it could be a good discussion to have with Codeberg people, I'm a member of the org actually (just regular contributor) and I'm sure people would be interested into blocking AI scrapping at the infra level.
# mathilde hacknorris: what should they have instead? A fediverse account? It's not better.
# hacknorris codeberg. neocities, etc…
# hacknorris none of it allow blocking ai as it's just impossible now
# mathilde everything public can be scrapped.
# hacknorris yup
# mathilde even myself hosting stuff on the very servers I own, OpenAI can still change their IP ranges without warning and then I'm back with scrappers going through my websites.
# hacknorris someone should make some other internet based on location i think. but it can't be me cause it'd need hardware i can't just buy without a big reason (still living with family)
# hacknorris plus most of things are blocked in most countries by law (waves)
mathilde joined the channel
# hacknorris we shouldn't use internet these times i think…
# hacknorris only such places which don't save data. like IRC
# mathilde hacknorris: it's not accurate
# mathilde IRC data can easily be made public, many services do it.
# mathilde look for yourself: https://chat.indieweb.org/2025-01-29#bottom
Guest6 joined the channel
# mathilde I personaly use Signal a lot, with social functionalities like stories it's very nice to keep in touch with people. End-to-end encrypted, ephemeral messages, open source, perfect. I wouldn't share anything confidential over the public internet (either my website, or IRC).
# hacknorris if only signal was natively for ubuntu touch in mobile version then yea…
# mathilde oh you are still using Ubuntu Touch? Amazing
# hacknorris yup
# hacknorris i do use
# hacknorris like maybe not exactly now but next to me is literally a samsung with u. touch
# hacknorris i could even join here from phone but i'm unsure if konversation'd allow me logging in (i have account)
nemonical and Hacknorris-mobil joined the channel
# Hacknorris-mobil Hiiii
# hacknorris yup
alfiee joined the channel
bterry joined the channel
# hacknorris what if i don't want to teach them?
# 
[KevinMarks] the indieweb wiki plus loqi has been an exercise in useful AI for a decade now. (I include previous hype cycles in AI to wind up generative LLM fanboys - it would count as an Expert System, which I think is 2 or 3 cycles ago)
# hacknorris people shouldn't have made internet…
# [Joe_Crawford] It would be great if people could distinguish sorts of chat bots. Eliza style (1967 tech) from infocom style games (1980s) from Markov chain based (2000s) from LLM based. The fact that we can’t certainly points to some kind of shortcoming somewhere.
# [KevinMarks] well, the redirects are a kind of chaining, as are the see also blocks - loqi is a Knowledge Acquistion tool implementing a prompting methodology
bterry joined the channel
# [KevinMarks] Reading https://en.wikipedia.org/wiki/Knowledge_Acquisition_and_Documentation_Structuring I think we could wrap loqi in similar jargon
alfiee, btrem, plantroon, JadedBlueEyes, mathilde, ttybitnik, bterry, aelaraji8, GuestZero, thegreekgeek and barnaby joined the channel
# carrvo[d] For blocking AI and scrapers, you would do better to try to block the user agent they advertise instead of IPs. Not a guarantee of course; but you could be rigid and only allow the user agents of known browsers.
# carrvo[d] I have also found a lot try a few starting points and then follow links so I have those places return 403 Forbidden. Unfortunately this includes the root of the domain.
# [Joe_Crawford] That's reminding me I never put on my todo list evaluating this ai UA list https://github.com/ai-robots-txt/ai.robots.txt
# carrvo[d] For phone services, I have heard that Canada is one of the highest because of how costly the infrastructure is (big landmass, tiny population). Ours start at $50CAD/month if you want any data. (You *can* be $20 for unlimited national and no data.) On the other hand I have heard that Africa is super cheap (cheaper basic essentials) and many use WhatsApp as a result.
# mathilde carrvo[d]: true, thanks for the advice regarding users agents. I think in the end it should be a mixture of different solutions.
# mathilde for me, it is much more convenient as a system administrator to block them on the firewall level, with IP addresses. It is more efficient than letting these requests being served by my reverse proxies or web servers.
# [aciccarello] This site is pretty useful for finding out what bots are scraping https://darkvisitors.com/
# mathilde [Joe_Crawford]: it is a good list, but far from complete.
# hacknorris aciccarello : what if someone doesn't have server?
gRegor joined the channel
# [aciccarello] I have a static site so I'm not actually looking at a server but the site was helpful when writing a robots.txt file even if many scrapers ignore it.
# hacknorris nothing abides by it except things which sholdn't
# [aciccarello] These insights are interesting. I didn't realize Amazon was so busy https://darkvisitors.com/insights
# mathilde +1
# mathilde here is mine, if you are curious: https://codeberg.org/mathilde/website/src/branch/main/robots.txt
# hacknorris only
# mathilde sebbu: you from fRance? Free customer?
# mathilde shame that only the 20€ plan includes switzerland
# 
Loqi robots.txt is a file used to inform web crawlers what parts of a site should or should not be crawled https://indieweb.org/robots_txt
# carrvo[d] mathilde unfortunately it is the more expensive firewalls that can sniff traffic to block more intelligently 😦
plantroon joined the channel
# hacknorris it is
# carrvo[d] Canada phone plan comparison: https://www.telus.com/en/mobility/plans (for those interested in differing prices across countries).
# carrvo[d] [edit] Canada phone plan comparison: https://www.telus.com/en/mobility/plans (for those interested in differing prices across countries).
# mathilde Free Mobile API is very cool, when you are a customer you can send SMS programmatically: https://codeberg.org/mathilde/handbook/src/branch/main/servers/exploitation/communication/freesms.txt
# mathilde Then, coupled at a basic checkdisk script, you get a free disk space monitoring: https://codeberg.org/mathilde/handbook/src/branch/main/servers/exploitation/checkdisk.txt
# carrvo[d] Sigh, this may have been better for #indieweb-random
# carrvo[d] sebbu, Canada is just introducing 2Gbps down...
plantroon and alfiee joined the channel
# Loqi Homebrew Website Club Europe/London is starting soon! Join us! https://events.indieweb.org/JqXSTx1zbNXy
# mathilde amazing!
plantroon, jak2k, jgee487, GuestZero and alfiee joined the channel
# mathilde Regarding the previous AI scrappers discussion: https://tzovar.as/algorithmic-sabotage-ssg/
nemonical and alfiee joined the channel
# [aciccarello] I like that idea. Punish crawlers that ignore restrictions.
d1, user23, alfiee, dpk, bterry, sebbu2, Oclair, ttybitnik, Zorb and plantroon joined the channel