Show newer
Shamar boosted

Folks, please do delete your X accounts, by all means, but I’d be very surprised if that meant that your data is deleted or that Elon Musk won’t be using it to train his AI, etc.

I put in a data deletion request after deactivating my account in 2022 and set the Irish DPC on them when they didn’t. After several back and forths, with the Irish DPC being less than worthless, I gave up. Afaict, there is zero enforcement of GDPR Article 17.

CC @noybeu

#x #twitter #gdpr #ElonMusk

What is the correct way to configure .htaccess to send all requests to a unix domain socket?

Ivtried variour enchantement with mod_proxy and mod_rewrite without success.

.htaccess only because I want it to work on shared hostings.

Also, what about NON php fcgi?
I couldn't even get it working with mod_proxy_cgi (again, .htaccess only). The receiving server is up and running, listening at the destination UDS.

Am I missing something obvious?

@gualdo

non credo che vendere cambierebbe il modello di business di , come avrebbe fatto 10 anni fa.

Se fossero costretti a vendere , invece...

Shamar boosted
Che una sede di pubblicazione sia scientifica per l'#ANVUR solo se si organizza in fascicoli, e che dunque Open Research Europe non lo sia può sembrare una bizzarria burocratica, ma non lo è. La #valutazione di stato, essendo #bibliometrica, ha bisogno che tutta la discussione scientifica sia marchiata e ingabbiata in riviste, e per questo motivo deve scoraggiare i ricercatori dal tentare strade alternative, perfino quando sono promosse dalla commissione europea.
Shamar boosted

Negli elenchi aggiornati delle riviste dichiarate scientifiche dall’ANVUR continua a mancare Open Research Europe (ORE), infrastruttura di revisione paritaria aperta offerta dalla Commissione dell’Unione Europea agli autori i cui lavori di ricerca sono esito di finanziamenti europei. Visti i precedenti, questa assenza non sorprende. Ed era anche già chiaro che la recente “Disposizione […]

https://aisa.sp.unipi.it/lanvur-e-open-research-europe/

@smallcircles

I started adding a POLITICS.txt to my projects 5 years ago, after writing my essay "What is Informatics?": encrypted.tesio.it/2019/06/03/

Writing the POLITICS.txt for a project is always pretty difficult: it means to be explicit with your hopes and ambitions while finding the right words to express them is hard and often lead other developers to mock you even if they use and appreciate the code you wrote.

You can see a recent example here: code.tesio.it/p/self-hosting/d

@jannem @abekonge

@jannem

I think nobody (except my daugthers) is entitled to my work.

Not even managers at work can force me to do something I don't want to do, so much that "It cannot be done" and "I don't know how to do it" and "give me a five years and a team of five very skilled engineers, and I'll do it" are well known show-stopper at work.

When I share my high-skilled work, I set up whatever license I see fit (sometime debating with board gatekeepers, cause I like they hate) and usually set an explicit POLITICS.txt to explicitly explain the socio-technical goals of the project.

Usually it does NOT cover governance because forks are welcome (within the rules of the license, obviously)

@smallcircles @abekonge

Shamar boosted

I attended the EDPB event on #PayorOkay models and left deeply concerned. The discussion lacked acknowledgment of data protection as a fundamental right and ignored clear GDPR principles making the model unlawful. Instead, it conflated ads with core services, sidelining fairness and rights. Surveillance ads harm individuals and society, yet their ‘value’ is overstated. We must reclaim the debate: data protection is key to human dignity and a rights-respecting digital future.

Shamar boosted

Bluesky and the Fedi: It's VC vs. DIY and you can't underestimate that distinction in your comparisons 

Bluesky received $35 Million in a year.

"The Fediverse" as a whole, may have managed to acquire a grand total of $500k in investment from individuals and nonprofits over the past eight years.

Mastodon itself has like 3 paid devs. None of them are getting rich.

Things just are not going to move at the same speed or at the same scale--and they literally can't, because the money isn't there.

It's the difference between VC-funded and volunteer-funded and any discussion or comparison to platforms needs to fully internalize that this difference does matter.

It's like wondering why the local bakery can't compete with, I dunno, BreadMax, the franchised international bread consortium with a store on every corner.

This isn't to say there aren't cultural differences, but ultimately it just feels like a much different conversation.

New person joins Bluesky and no one really talks about the infrastructure because you can't just "host your own," not really. Hosting your data isn't the same as hosting the "means of posting and discoverability itself." Moderation isn't outsourced in the same way as it is to every individual server admin here.

Also, a person joining Bluesky can be like, "WTF devs?" and devs who are literally working fulltime, every single day, with no side projects or silly distractions like "needing to support themselves with a job to live" to take away their focus from building the thing, will build the thing.

And "the thing" will materialize, as if by magic. Wow, when you fund a development team and dedicate them to a task and when the millions fall from the fucking skies, it's amazing what you can do!

Isn't it, though? If you put money behind a dev or design team, suddenly stuff starts happening.

The problem comes in trying to look over here at Johnny's little Snac activity pub server and mocking it and going, "Damn, Snac, get it together. Bluesky has quote posts already. And where's your reply control?"

And the Snac dev/devs (I am just picking one project at random) look up and go, "Uhhhh, I don't actually WORK for you. Evidenced by the fact that you aren't paying me a goddamned thing to work on this, so I'll get to your concerns eventually, but watch your fucking tone or make your own or just fuck off, I don't care."

And people will go, "Damn, this place is rude."

Well.... YOU are kinda rude too, you know? I even do this to some degree and it's not okay.

Like when I complain about Mastodon, I'm basically demoralizing a (largely unpaid) team who is drastically underfunded for the scale of projects against which they are competing.

We just don't have millions of dollars to fuel development. Lots of times, it's people building the shit for free. So I get that you're mad that we're not farther along, and that stuff moves slowly, and whatever, but it's hardly a surprise that a team with access to millions of dollars is moving more quickly than a space with development funded largely--although not entirely--by people working for free.

And, I dunno, when you come into the house of people working for free to build a thing and you yell at them about the thing, it hits a lot differently than when you're yelling at a company with millions of dollars in the bank and the resources to do the stuff you want at the speed of the market.

@grunfink

I think that with a pid and an oprion to automatically demonize (sysv way, doible fork, setsid, detaching from the terminal...) it would be trivial to run snac on a shared hosting even if it doesn't support fastcgi: you just need a script to start it.

@lrvick well, I think it's unfair to compare to the standards, given how poisoned all of them.

@grunfink

Thanks, I've seen handles SIGINT by terminating politely.

does it save a pidfile somewhere?

I can't find any in the data/ dir and I can't find an entry in the server.json

@grunfink

Hi, I managed to build a statically linked with musl (and zlib, curl and openssl).

Now I realized that I could deploy it as a FastCGI, but I wonder how I could stop a running process on snac update.

Is there a kill switch one could toggle eg by creating an empty file with a specific name under the data dir? Or maybe a version file that is periodically checked by snac processes that exit() when you start a new version?

Or maybe something even simpler I'm missing?

Also, I've read that snac now support unix domain sockets, but I can't find anything about that in the doc... where should I look?

@stefano

Shamar boosted
I'm glad to announce the release of version 2.63 of #snac, the simple, minimalistic #ActivityPub instance server written in C. It includes the following changes:

The server can now act as a proxy for all image, audio or video media coming from other account's posts (both from the Web UI and the Mastodon API). This way, other servers will see media requests coming from the server IP, not the user's, improving privacy. This is controlled by setting the proxy_media boolean field to server.json to true.

The strict_public_timelines option introduced in the previous release now works correctly.

Fixed a crash when posting from the links browser.

Fixed some repeated images from Lemmy posts.

Fixed a crash when posting an image from the tooot mobile app.

Updated FreeBSD rc script: the server process is now managed by the daemon(8) utility (contributed by @stefano@bsd.cafe).

RSS feeds are now in 2.0 version instead of 0.91.

https://comam.es/what-is-snac

If you find #snac useful, please consider contributing via LiberaPay: https://liberapay.com/grunfink/donate

#snacAnnounces

This release has been inspired by the song New Moon (Dark Phase) by #DuranDuran.

Interesting library @rysiek

Have you considered a browser plugin?

@nlnet

@tomgag

It's not.

It's a stance OSI got by its sponsors opensource.org/sponsors

Meta in particular was allowed to cancel other teams' votes.

The issues have been reported by several people for months (here a recap discuss.opensource.org/t/list- ) but OSI ignored them, silenced or censored those people.

See Julia Ferraioli's account of the co-design process: linkedin.com/posts/juliaferrai

If you feel disgusted as I do, here an initiative to restore (and _maybe_ improve) the OSD through public discussion and IETF rough consensus opensourcedefinition.org/wip/

Shamar boosted

@knowprose @sj Thanks for sharing this, it's good to know there's a sizable community that is unhappy with how OSAID got rolled out.

Shamar boosted

@sj @jaredwhite Through LinkedIn, I got invited to this discussion area regarding the topic... So I'm passing it along.

I haven't had enough contiguous time yet to chime in, but I will.

*Dusts off penguins and Gnus*

discuss.opensourcedefinition.o

@phab

trovato.

per dar peso a Musk, non hanno proprio niente da fare

@cescobarresi

Show older
Qoto Mastodon

QOTO: Question Others to Teach Ourselves
An inclusive, Academic Freedom, instance
All cultures welcome.
Hate speech and harassment strictly forbidden.