code51

> Since you are paying, you can contact support and actually get a human to help you.

Wow, a rare offering from Google.

> "Twitter’s newsletter service Revue will be shut down by the end of the year. Revue was acquired by Twitter in January 2021 for an undisclosed amount. The report also says that Notes, a longform posting feature, is on hold, as is Twitter’s plan to build crypto wallets."

All these are look so distant to Twitter's core value.

They say it was announced through company-wide email so I think pretty easy to be confirmed in around 1 hour by the media.

I appreciate the idea and implementation.

However this is literally GQ magazine implemented as a search page which overwhelms the user with irrelevant information.

That's why old Google was cool: clean and relevant. I don't see anyone looking for efficiency (not social media but a search page) would want to change that.

You're right, this person does not represent Google but here's the hard truth:

Google does not offer Adsense/Admob support for ordinary people. No way to reach them.

a) Unpaid balance could just be refunded to advertisers.

b) Worries about child labor: already happened. Google profited from his activity already. It's just the kid not receiving anything.

Google Adsense/Admob offers no support for most of its users.

Only users that generate large revenue can access Chat option.

For the rest, the only option is to write to support community. It's deliberate: feels like there is support but there is none.

I think for this to work, Google actually has to delete any information about you completely from its systems. I'm curious if anyone tried this with success as well.

If Google was so worried about this, their AI and behavior analytics can %99.9 know you are underage. They could show a BIG RED alert multiple times about this. Plus, I don't see heavy upload traffic from these young people becoming a liability on their end which is basically free work.

a) Imagine you are 13 years old and actually reading Terms of Services. Just imagine you are under 18.

b) It's legal speak mentioning "you cannot create a new account" but it doesn't say you can't try to fix the problems about your existing account. The problem is there is nobody to reach for mistakes or downright violations from Google's side. It doesn't work to help people with good intentions realizing their errors.

c) It's not mentioning that you will forever be blocked because you misrepresented your age for whatever the reason (just an unaware click is enough).

If Google was actually serious in enforcing the policy, they would just check any detail that could trigger this user could be below 18 years of age. As you read here, they don't spend much of an effort to avoid this scenario. AI works just to block you, not to avoid your downfall.

d) I wonder Google informs kids this much before making money on free content they upload on Youtube. It's not hard to see this is one-sided exploitation in business sense.

So... pretty clear, yeah.

A 13-year-old boy sets up Adsense for his Youtube channel and his father is told that his son is now banned from Adsense/Admob for a lifetime.

This is what you get when you have no humans in the loop.

> next year as the downturn ends and the upswing begins

Who says the downturn ends next year?

VCs act like it's all 2008: someone will come and bail them out. Yet, we are literally one Powerpoint away from a multi-year dumpster fire.

First off, it seems that the model can easily run on M1/M2 with minor modification. However `aten::_index_put_impl_` operator is current not supported and fallback always slows things down quite a lot.

Second, is there a bug with how the script processes incoming audio segments? For a short 4 second clip, what I got was:

> [00:00.000 --> 00:03.760] Okay, Eunice, travel plans. I need to be in New York on Monday, L.A. on Tuesday, New York on Wednesday, L.A. on Thursday. You're knocking Friday. Got it?

> [00:03.760 --> 00:28.760] Got it.

However the final segment should have been shy of 1 second. It mistakenly thinks the last segment was 25 seconds long and makes you wait for processing.

That's why petrodollar is part of heated conversations around the world.

You do this enough and your allies become collateral damage, the system begs for a change. Rest of the world starts to only base their trades in gold (or some other standard) and when faced with USD debt, change the debt amount through policy. US has to keep their (future) allies afloat either openly or with hidden support.

There are large crowds that took 2 doses in pandemic period (getting vaccinated properly back at the time) but not receiving boosters.

It would be easy to track:

- 2 dose, no other booster

- all boosters

- no vaccine

and readjust results considering age and other bias etc.

If all booster people have excess deaths, than "not from Covid" ? If 2 dose people excess deaths < all boosted people, than something about vaccine? No vaccine could be control.

You cannot effectively "compartmentalize" when it comes to Google. It actively tries to resolve you into one person, most of the time it succeeds. Or worse, you are marked with a false positive.

I've been following your MPS branch and have run it but couldn't address the issue without this explanation. Thank you!

Without k-diffusion support, I don't think this replicates Stable Diffusion experience:

https://github.com/crowsonkb/k-diffusion

Yes, running on M1/M2 (MPS device) was possible with modifications. img2img and inpainting also works.

However you'll run into problems when you want k-diffusion sampling or textual inversion support.

The trouble is that these positions try to filter you out asking for the same concepts in interviews. Nobody cares for how good a programmer you are. Data engineer positions don't work well if you want to touch cutting-edge concepts like big-scale DNNs. Sure you'd learn about deploying things but if your day-to-day job is not about e.g. ONNX, pytorch, JAX and massive scale training/inference pipelines, then you'd still be left out.

Chicken and egg problem. Probably the optimal way is just aiming for top ML research teams right from the start as a new grad. No other way is better.

I'm not based in US so even contacting a lawyer about this thing is expensive.

Lawyers here are not willing to handle international cases except big business cases.

6+ years... wow.

I'm afraid I'll end up like this:

- unable to use Adsense/Admob for years

- and not getting the money

- money becomes basically nothing because of inflation

I think the only letter the legal team expects through google-legal-support@google.com is a notice of sueing them. I really don't have the power to sue them in a US court.

Do people have success when they notify (not sue) the legal team about Google acting against their Terms of Service?

Exactly.

Google is not paying my Admob money even after 5 months has passed (no violation, self-close of account)

Through the Kafkaesque maze of non-existent Adsense support, my case (they say) finally got escalated to a payment person at Google. Yet, I didn't hear update for the past month.

You can't ask anything about your case and you can't get any time projection on when you'll hear anything back.

If you don't have big traffic or huge social reach, you are expendable and can get stuck in a never-ending hell.

Here you go: https://github.com/facebookresearch/fairseq/tree/nllb

Plus, why would an advanced civilization that has already understood the value of energy waste this kind of energy trying to contact us?

- If they are advanced (+civil), then they would have seen a critical development that this became a necessity, rather than preference.

- They may not be civil: not caring what energy they waste. Advanced but not responsible beings.

- They may be doing something for themselves, leaving traces to different directions. We may have stumbled on a construction site. Nothing of value for sending a response. Would you beam a response to reaction site of a supercollider? In case you did, would you get a "message" back? I doubt it.

- Most probably: interference.

> 4) Someone who is sick or is recovering from illness is unlikely to get to the gym early.

Someone somewhere just said "Challenge accepted!"

If the smallpox vaccine has not changed a lot in the meantime... Probability of death: 1 in 1 million. Probability of death for infants: 5 in 1 million.

Still not something to take a risk.

[Comparatively, if this source is not mistaken: https://www.cdc.gov/coronavirus/2019-ncov/vaccines/safety/ad... for Covid vaccine 5 deaths per 1 million]

UBI first then automate? or Automate first, then UBI? Massively different options for millions of people.

The AI dream we dreamed of is actually this kind of nightmare in ignorant hands.

If you are a false positive for Google bots, then you're toast.

Your ad income will vanish, you app will be forgotten and even your very email and associated data underneath is on the brink of lifetime ban.

This is the very essence of Google. If you are big enough as Google, you can swing a huge f* you gesture to the whole world of software developers.

"Don't be evil" - yeah right.

The very first thing Google makes use of their fat cutting-edge language models is banning ordinary people from the whole ecosystem for their tiniest fixable mistakes.

Google scales evil before they scale good. Always have been, always will be.

We as little software developers don't have any significant resource to sue the hell out of Google and they can get away with this pure evil every single day.

For non-English locales, Google sucks 99.9% of the time for popular queries since they allow spamming from big media outlets. A top result is literally composed of spammy SEO keyword content filled with a chain of questions in the article. So (I think) it sucks for 'normal' people in non-English.