lerela

https://archive.ph/KuvYm

Randomly clicking on the map led me to learn about this: https://en.wikipedia.org/wiki/Laconia_incident

Would love to be able to filter for events with a Wikipedia page (assuming they were the most significant)!

Thanks for the report!

We have clarified the documentation, sorry about the confusion! 16GB should be enough but it requires some vLLM cache tweaking that we still need to work on, so we put 24GB to be safe. Other deployment methods and quantized versions can definitely fit on 16GB!

Does no one else find it disturbing to publish the identities of these public employees with their salaries? Even if the data is freely available somewhere, this is high visibility. Would the article have been any less interesting if their names were anonymized and their privacy slightly respected?

I find that publishing the names and salaries of random S.F. security guards is quite an unsettling level of transparency.

A few years ago, the package I received contained a bottle of water instead of the phone I ordered. Amazon promptly refunded me and I was quite pleased, because I was a bit concerned by how I would sound to the customer service representative telling this story

On my end examples do not work (empty placeholders) with Firefox (on Linux). They work well with Chrome though.

The FP16 7B version runs on my Ubuntu XPS with 32GB memory, ~300ms/token. 13B also works but results aren't really good (the model will loop after a few sentences) so parameters probably need tuning.

So far I'm unable to reliably generate outputs in a different language than English, the model will very quickly start to translate (even if it's not asked) or just switch to English.

Absolutely, I meant that the app is not used to collect the email address since it is already in our system but it is true they have no way of knowing.

Funny coincidence that I have received a notification from the Play store this morning saying that analysis of my app (with super tiny outreach) showed that my Data Safety Form was inaccurate and asking me to fix it asap.

Seems like letting the user enter an email address to recover his password counts as email collection (he cannot subscribe, provide or see his email using the app, hence the initial form I submitted).

Basic plan is $35/m which is much higher than alternatives discussed in this thread, isn't it?

From the article:

The main reason why they don’t care about serving users is because the user is not the customer at Facebook. The users (you and me) are the product sold to the actual customers.

I'm curious: is the "please" you added purely a communication habit or does it lead the model to provide better results?

It's the same with Gitlab issues and comments and it's a small but very real pain.

But many people (okay, me at least) just want to watch one or two videos once a week, so a subscription makes no sense... The article also mentions how absurd un-skippable ads have become. The whole "casual YouTube watching experience" is a real pain now, and I do not know if focusing on hard-core tubers will prove right in the long run.

Latest models have suspend issues with Linux though, so keep that in mind if you do not plan to use Windows.

I'm also doing this and see multiple benefits.

However I've recently been bitten by my catch-all, using a money transfer service with the email worldremit@mycatchall.com (guess the company). When they asked for additional documents to verify my account after many months, they never received my reply and I ended up banned. I could not login anymore. When I reached out from another email address, they refused to process the documents because they originated from another, unauthorized email address, and asked that I resent the original email from the registered email. I suspect their anti-phishing filters just ban any email containing "worldremit", so it never got through and despite multiple thorough explanations I could never get someone to listen or reinstate the account.

I'm still getting the newsletter though, because unsubscribing requires logging in first... But then I can just ban this email address, so at least the anti-spam strategy works!

I am surprised you are saying K9 does not have a unified inbox, because I use it daily (Settings/General Settings/Display/Show Unified Inbox). Do you mean that the way it works does not meet your needs?

As others have pointed out, $19/user/month is just not an option for many small teams, especially with non-western costs.

I guess we'll start planning the migration but it's really disappointing not to be able to pay a reasonable fee and then be kicked out on a 3-months notice.

I've seen complaints about this business model for a while on HN but never got the chance to understand Gitlab's reasoning for sticking to this pricing.

Moreover it is quite disappointing to see that Africa is one of the most expensive region even though the purchasing power of the population is much lower than the rest of the world.

Not sure how to build competitive services for that region based on that pricing.

Tough timing, considering their IPO is tomorrow [1].

[1] https://ipo.ovhcloud.com/sites/default/files/2021-10/OVHclou...

Only reason I upgraded was to get 32GB of RAM. So far, it's underwhelming...

On the latest XPS that's not even an option: https://www.dell.com/community/XPS/Ubuntu-deep-sleep-missing...

(I've just seen the last posts on that thread and will give it a try)

I have an high-end XPS 13 from 2020 (9310), allegedly with native Linux support, and have the exact same issues.

My previous XPS 13 from 2016 was suspending properly but now I cannot suspend my "laptop" for more than a day (it will die) or store it in a bag (it will become lava). Hibernate does not work either.

I'm learning with this thread that it also happens on Windows and I'm struggling to understand how Dell could decide to sell "laptops" at this price and not test one of the most basic features of a laptop.

Despite those difficulties (that do not sound ludicrous considering the complexity and the amount of technology transfer that was required, but I'm no submarines expert), as I mentioned in another comment, the joint communique between French and Australian Defence minister from August mentioned that "both sides [were] committed to deepen defence industry cooperation and enhance their capability edge in the region" and that "ministers underlined the importance of the Future Submarine program".

Looking back, it is hard to argue that official line was not very disingenuous, even after acknowledging the issues with the program.

French and Australian Defence ministers met in August and issued a joint communique [1] mentioning that they discussed "enhancing and diversifying France’s military co-operation with Australia" and stating that "both sides committed to deepen defence industry cooperation and enhance their capability edge in the region" and most specifically that "ministers underlined the importance of the Future Submarine program".

Announcing 2 weeks later that the contract was cancelled sounds like a "sudden" announcement to me, while the press reports that discussions had been ongoing for almost 18 months in secret make that communique slightly deceitful from Australian government imo.

[1] https://www.foreignminister.gov.au/minister/marise-payne/med...

After a few tests it looks good but it reports some domains as registered when they are not (for instance kostikaable.com or kostikaition.io are marked registered but do not seem to exist when I manually run a Whois). Maybe the tool is running onto some rate limiting thresholds?

Also when there are a lot of domains to check you could check them in parallel, otherwise it gets quite slow to wait for the results (but that might be an inherent limitation of the Whois servers).

Unimportant but it is a bit misleading coming from a Linux browser to only see Download for Windows & Mac call-to-actions. I thought it was incompatible before I saw the top right penguin. You might wanna use the User-Agent to show a more relevant CTA.

Not a feature per se, but being in the process of building a PWA for iOS with access to the camera, the fact that the authorization dialog pops up every time the app is opened (meaning that the permission is not saved even if the PWA has not been modified) looks like a deliberate attempt to make this technology not truly viable, because it really provides terrible UX.

I've come to really like peek since it allows to select the rectangle you want to record with minimal interaction. It is also in the Debian repos.