Right, the other thing i considered is that you could just create a company and “buy” the data from them for a ridiculous amount of money and then you have less requirement to detail the data. Similarly you could deem the data unsharable and fudge the provenance.

Like locks, it will only keep honest people honest.

the actual license text part being questioned .

Data Information: Sufficiently detailed information about the data used to train the system so that a skilled person can build a substantially equivalent system. Data Information shall be made available under OSI-approved terms.

In particular, this must include: (1) the complete description of all data used for training, including (if used) of unshareable data, disclosing the provenance of the data, its scope and characteristics, how the data was obtained and selected, the labeling procedures, and data processing and filtering methodologies; (2) a listing of all publicly available training data and where to obtain it; and (3) a listing of all training data obtainable from third parties and where to obtain it, including for fee.

(The rest of the license goes on to talk about weights, etc).

I agree with you somewhat. I’m glad that each source does need to be listed and described. I’m less thrilled to see “unshareable” data and data that cost $ in there since i think these have potential to effectively make a model not able to be retrained by a “skilled person”.

It’s a cheap way to make an AI license without making all the training data open source (and dodging the legalities of that).

+1 for gitlab. You can programmatically generate a csv file that can be used to generate issue(s) which support markdown format. Then your checklists could be issues and marked as completed when done.

You could also for instance set up a weekly pipeline schedule to generate issue(s) from the csv if some of the issues are needed on an interval.

If gitlab isn’t an option then id still look into generating the .md files this way and finding a home for the .md files that works for your user(s)

There are much smaller projects that ask for more from commits/merge messages. This is a normal ask

Wait, if they suspended your domain, can you even transfer it away? if not, that’s really fucking scary.

Njalla takes ownership of every domain purchased on their platform. They do let you transfer domains to another registrar where you could be the owner if your account is in good standing but seems like that may not be the case here (since account suspended)

That may be great for some domain use cases but for most stuff it would be better to have your name on the domain registration

Gitea, took control away from community and gave it to a for profit organization. Forgejo was born

Python 3.13 is adding support for removing GIL, via PEP 703

I really like that it is a static website being updated and built on a schedule from github actions.

From what i can tell there are no transaction fees for sponsorships from personal accounts, and organizations pay 6% (or 3% if invoicing). (Source)

Is there something else I’m not seeing?

The whole idea to check the donations came from stumbling upon this post which discussed costs per user. Even $1/mo is quite a bit more than the average user cost. So $2 isn’t so measly when putting it into that perspective!

In fairness websites from 2000-2004 werent all that better

Were there better ways to make a site? Absolutely, but it is much less wild than if you told me that this happened last week. Plus i would hope they were just churning out websites for cheap since a lot places didn’t have a website, or they used geocities/similar

It would be much more customer and developer friendly to allow linking a service portal instead of providing a phone number. I would go insane if a user called me directly every time one of my projects had a bug or some perceived (non)issue. No, that’s not how this works.

Never. For the most part i haven’t had a question that hasn’t already asked or that couldn’t be answered from reading the docs or some other source. For the cases i get stuck i ask the question to a more focused group

And if you want a private repo, you can also use gitlab and point to custom domain with gitlab pages or cloudflare pages.

Yes, oracle will reclaim your server if it falls under certain thresholds for the resources you’ve signed up for. So it might be better to request less resources then you need but this will somewhat complicate things if you want more resources in the future since iirc you can’t simply resize.

One way to get around all of this though is convert to pay as you go (PAYG). PAYG gets the same always free allocations and you only pay for use above that, and oracle won’t reclaim PAYG (at least not my server for ~4 years). Just set up a budget of a $1 and then alerts to email you if you reach 1% of your budget. If you somehow go over your free resources it’ll tell you.

Lastly in some cases oracle just straight up loses your data or disables your account. As always practice 3-2-1 backups (don’t rely on the free rotating backups on their servers as your only backup).

It’s some hoops to jump through but i was paying $5/ month for a digital ocean droplet and the oracle server has been running for 4 years now, and i also have scaled up one project and started a few others that wouldn’t have all fit on my droplet. Other than the threat of reclaiming my resources before i switched to PAYG I’ve been pretty happy with it.

The first part of the article talks about how to use git notes and has an example commit, followed by adding the note, and then viewing the note. This is all native git.

The “problem” is that we have centralized discussions in github/gitlab comments and if we want to retain that data then we need to convert the comments into gitnotes. The CLI part is that specific discussion on how Symfony uses git notes to store github comments. It references an internal CLI but then goes through an example of how to use github api to fetch the comments, create git notes, then push those git notes to github. So while the symfony CLI is internal, it looks like we’re given an example of how to do this for github.

Ive been meaning to move to codeberg, self hosted forgejo, or sourcehut so this will only accelerate that if things get worse.

Thanks! I’ll add that to my list to check out

If you only ever keep your repository private AND it is not a fork of a public repo, then you are fine. Full stop.

If you ever fork the repo and make a “INTERNAL” private fork but move the main project public then anything you commit to the private fork will be discoverable through the public project.

Basically you should assume if you make a repo public then the repo and all of its forks will be public-- even if the forks are “private” the commit data can be found through the main repo.

The original report: https://trufflesecurity.com/blog/anyone-can-access-deleted-and-private-repo-data-github