Hacker Newsnew | past | comments | ask | show | jobs | submit | mimimi31's commentslogin

>Isn't this suitable for a Bayesian classifier?

I think that's what GnuCash does by default. Even with years of past transaction data it still gets some very obvious matches wrong for me. In my experience it's about 90% accurate for the ones it really should be able to do based on the training data.


> "...it's about 90% accurate for the ones it really should be able to do based on the training data."

What's the pathway for the remaining 10%? Are they simply misclassified, and dropped into a queue for manual labeling? Do the outliers get managed by the GnuCash? Or do they get dumped into a misc 9000 account?


It shows you the automatic account matches on import, allowing you to double-check and correct any misclassified ones.

Ok. So what you're pointing to is not an automated pipeline, but a user mediated process. It's the same pattern in QuickBooks, or whatever ERP.

>There is a sign-in/create account screen, but this is easily bypassed if you know how to edit an sqlite file.

You can also just add "SideloadedMode=true" to your "Kobo eReader.conf" to achieve that. This removes the "Home" and "Discover" tabs as well, defaulting to the clean "My Books" tab instead.


>Add special signals you can change on your server, which the app will understand, such as a forced update that will install without asking the user.

I understand the reasoning, but that makes it feel a bit too close to a C&C server for my liking. If the update server ever gets compromised, I imagine this could increase the damage done drastically.


I tried the the linked example search for "baby cat" and it returned the same three AI cats you can see in their Google search comparison screenshot on the first page. None of them labeled as AI generated.

Edit: When I explicitly choose to "Include" AI images from the toolbar option, they disappear. When I choose to "Exclude" them, they reappear. Still seems a bit buggy.


>I don't remember where I've used my YubiKey in the past.

I track this in my password manager. Accounts where the YubiKey is enrolled are tagged "YubiKey (FIDO)".


This would probably be a good place to suggest to others here to track which accounts you've logged into via Google or other social media oath.

I just had to log into stack overflow for the first time in years, and did not remember what I used to previously log in. Once I figured it out that information went into Keepass too.


Just fyi, Google lists sites you’ve used to “sign in with Google” in your account setting page. Apple and GitHub have this as well.


You should assume Google, GitHub, and Apple are hostile and try to limit your blast radius. If you have an account problem they have no customer service to help you.


I can't get into my Google account that's almost 20 years old because I only have the username, password, recovery email and have all the email forwarded to me, but I no longer have the phone number and they silently enabled 2FA SMS at some point.


I wonder if you can phone the phone number, explain the situation and offer to venmo/paypal the new assignee money for the 2FA code

You could try every 3 - 5 years or so as it gets reassigned again


I tried this before, but I haven't tried for a couple of years -- you're right, it might have got reassigned -- I will try again!


+1


Yes, this is exactly why I won’t use these federated identity features of platforms like this. I have a reasonable amount of trust that they are mostly secure, but I have zero trust that they will be helpful if I ever have account troubles. What I don’t need is to have Google (etc) auth problems cascade down to every other account I own.


What's the alternative? What do you use for your base "recovery account" email?


My own domain’s email, which can easily be forwarded wherever I want.


Very well stated, concisely. Thank you.


Thank you, I should review that.


1Password has this functionality and it's excellent.


I really wish Bitwarden had more robust tools for organizing, sorting and tagging passwords. The current system of sorting them into folders is practically useless.


I don't get those prompts with Google Photos. Have you tried selecting "Use without an account" in the account menu at the top right?


Thank you, I didn't even consider this to be a possibility. I back up to my own storage and was annoyed by this message.

Untying photos from my google account is even better!


>Not if this is one of a few dozen or few hundred similar ongoing operations. The risk is always there, they have to expect some amount of failure.

That actually makes me think it's not happening at a larger scale, since we'd likely have heard of at least a few similarly elaborate cases being uncovered by now. If not during the attempt itself, then at least at some later point in time.

Either almost all of these operations remain undetected, because they are even more sophisticated and much of the world's software ecosystem has been secretly compromised for years or there aren't actually that many such operations.


>OpenWRT's hardware table seems not very useful; I can't, for instance, filter for at least two ethernet ports.

You could download the CSV dump of the hardware table [1] and filter in e.g. LibreOffice Calc.

[1] https://openwrt.org/_media/toh_dump_tab_separated.zip


Last time I checked the Linksys E8450/Belkin RT3200 was widely recommended.


The Linksys E8450/Belkin RT3200 is a solid and affordable router. UBI firmware support, hardware NAT offloading and DSA support for better VLAN performance. Less than perfect WiFi range.


If I was buying today, I would want something with more than 1GbE.


That wouldn't work for dynamically injected ads that could be at different timestamps and have different lengths for each user.


Some sort of checksumming to detect segments differing between users would probably be doable.


I think you under estimated the cost of your solution, the cost must not excess the profit from the ad.

You need hard work on the encoder to do that (at least to segment video, because re-encoding dynamically is obviously not an option). Not profitable for Google.


Aren't there codecs that don't carry state across keyframes? Wouldn't it then be trivial to split a video at a keyframe and insert new content?


Sounds like it would be just as trivial to detect the split at the decoder level.


Why? How would you determine if the content that comes after the split is an ad? What if YouTube has 1000s of versions of the same ad, of which they insert one after the split?


Shazam-style audio fingerprinting can help.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: