More importantly, the current dev branch source for packages/opencode/src/session/summary.ts shows summarizeMessage() now only computes diffs and updates the message summary object; it does not make an LLM call there anymore. The current code path calls summarizeSession() and summarizeMessage(), and summarizeMessage() just filters messages, computes diffs, sets userMsg.summary.diffs, and saves the message.
While I don't have OpenClaw installed and not sure how I 'd use it I doubt all the hype around it is because it doesn't solve a real problem. The project grew to huge popularity organically!!!
How can that happen if it doesn't serve a need people have?
NFTs were fueled by two different drives. One interested in the technology and if it could do something new and interesting, and another seeing it as an area of speculation (be that fueled by get rich quick and cash out or thinking it is a long term investment generally driven by how much the first factor played in).
OpenClaw seems to lack the monetary interest driving it as much. Not to say there is none, but I don't see people doing nearly as much to get me to buy their OpenClaw.
So, yes, on some level, hype alone doesn't prove use, because it can also be because of making money. But, on the other hand, the specific version of hype seems much more focused on the "Look at what I built" and much less on "Better buy in now" from the builders themselves. Of course the API providers selling tokens are loving it for financial reasons.
"And that's why we've created MailCoin, the best way to perform stochastic mailbox ablation with with the latest, hottest blockchain technology." - from Show HN, March 20, 2026
I'll ignore the bait and answer: NFTs were gambling in disguise, these claws are personal/household assistants, that proactively perform various tasks and can be genuinely useful. The security problem is very much unsolved, but comparing them to NFTs is just willfully ignorant at best
I agree, this is clearly an indictment against LLMs. If LLMs and agents were capable they'd 100% write it natively but they realize the current limitations.
I have tested this and they are very capable, replacing the modern windows calc (38mb mem) with an identical app (ux, looks, features, accessibility, localization) ended up with an app using 2MB memory.
You just need to point it in the right direction (c/rust/go etc) and be harsh with the requirements, especially memory usage.
If I was Microsoft I would use AI for this rather than badly embedding AI everywhere, a lot of power users would be overwhelmed by a win 11 update where OS apps/features dropped mem usage by 90%+
There is, and it would cause an outage while still not achieving the supposed goal of not going over budget. You don't want to be killing your customer's production over potential misconfigurations/forgotten budgets. Especially when you'd continue to bill them for the storage and other static things like IPs.
It's so much easier for them to have support wave accidental overuses.
Is it really re-writing - legally - if you are starting from the codebase itself? Not a lawyer, am wondering however if the Google vs Oracle Java lawsuit has some implications for this.
Haven't Google stopped publishing device-specific trees, hardware definitions, and drivers in AOSP so newer pixel are not that easy to use in GrapheneOS anymore?
It was a complicated product that many people worked in order to develop and took advantage of many pre-existing vulnerabilities as well knowledge of complex and niche systems in order to work.
Yeah, Stuxnet was the absolute worst of the worst the depths of its development we will likely truly never know. The cost of its development we will never truly know. It was an extremely highly, hyper targeted, advanced digital weapon. Nation states wouldn't even use this type of warfare against pedophiles.
More importantly, the current dev branch source for packages/opencode/src/session/summary.ts shows summarizeMessage() now only computes diffs and updates the message summary object; it does not make an LLM call there anymore. The current code path calls summarizeSession() and summarizeMessage(), and summarizeMessage() just filters messages, computes diffs, sets userMsg.summary.diffs, and saves the message.
https://github.com/anomalyco/opencode/blob/dev/packages/open...
reply