I would suspect that some of the slowdown that the author encountered does occur with even a dozen or so add-ons. Why else would Firefox bother you about resetting your profile if you haven't returned in a while?
Judging from the side it seems like there aren't a lot of developers and the current few have their favorite subsystems, those get almost all of the attention. The rest is kept as-is and does not progress. I also don't know if there's a trivial way to find how many external contributions they get from BugZilla, if they even get any?
I've tried so many while doing a bit of research into MUAs and I can't say there are any alternatives that could replace it properly. But depending on your usage or if you're not a demanding user, there might be something out there.
> If you don't click that last button, outlook uploads your IMAP email credentials to their own MS Cloud instance, and that proxies all your emails via microsoft's cloud servers. Do they read your email messages for advertising? Nobody knows!
I've seen cases where people have it set up like that and it's so awfully slow. Minutes to display a single new message. That cloud brings absolutely zero user-benefit.
I saw a tweet saying that there's a requirement for verification.
> Effective October 16, 2025, Microsoft will initiate mandatory account verification for all partners in the Windows Hardware Program who have not completed account verification since April 2024.
> Partners who fail to complete Account Verification by the deadline, or who do not meet the requirements, will have their status set to Rejected and will be suspended from the program.
I feel like the maximum effort mode kind-of wraps around and starts becoming "desperate" to the extent of lazy or a monkey's paw, similar to how lower effort modes or a poor prompt.
I think over-thinking is only solved by thinking more, not less. This is only viable once some intelligence threshold is reached, which I think Anthropic has borderline achieved.
> I think over-thinking is only solved by thinking more, not less.
Despite "thinking" tokens being determined by the preceding tokens, they still are taken from some probability distribution, just a complex one. This means that at each token selection step there is a probability P_e of an error, of selecting a wrong token.
These errors compound exponentially: the probability of not selecting wrong token for N steps is 1-(1-P_e)^N.
The shorter "thinking" is, the less is the probability of it going astray.
> The shorter "thinking" is, the less is the probability of it going astray
As long as the error introduced by more steps is less than the compounding error of sub-optimal token sampling, I would expect a better result.
I think your choice of "wrong" is extreme, suggesting such a token can catastrophically spoil the result. The modern reality is more that the model is able to recover.
> The most concrete issue for me, as highlighted by djb, is that when the NSA insists against hybrids
The fact that only NSA does that and they really have no convincing arguments seems like the biggest reason why the wider internet should only roll out hybrids. Then possibly wait decades for everything to mature and then reconsider plain modes of operation.
reply