Overall comment about your mail: I think such threads are *asking* to cause a flamewar. Pick lots of random, unrelated stuff, and complain about its state - the "lots of" guarantees that you'll make as many developers feel affected negatively as possible. So you'll have lots of stressed people arguing, which is the recipe for a flamewar.
I do realize that you do this out of a positive intent, but IMHO it is the same pattern as all the other flamewars we've had some weeks ago, and they were highly destructive and demotivating. Do you realize that we've had 30 pull requests during the past week? We don't need big motivational mails, people are churning out insane amounts of code :) If you'd like something such as darknet invites fixed which has been annoying you for years, it might really be the time to start writing the code :) Sorry. I'll try to do my part with fixing WoT, see below. On Monday, January 04, 2016 02:26:28 AM Arne Babenhauserheide wrote: > - WoT consumes too many resources (build 18 is faster, but my node > OOMs now, also without Sone). What is the memory limit? Can you analyze this with VisualVM, i.e. see which classes are using the most memory and give me a screenshot? > recover LCWoT and LCIntro and activate them by default Not necessary: Even though I'm a volunteer, I'll focus on working on fixing the biggest remaining performance issue of WoT for the following months (once I'm finished with the huge IRC/mail backlog of the past weeks). I churn out something like 3 hours/day currently; albeit it's probably gonna reduce to 2 h/d once the holidays are over. My work will includes fast startup, and reducing the default amount of USK subscriptions from 13 000 to something like 5. Also, I think it would be a bad idea to ship LCWoT: - It's faster due to the hack of running trust computations only once every 3 hours. WoT updates its output immediately whenever a trust value changes. This means that the LCWoT output will be *incorrect* most of the time compared to WoT. For example, reacting to a spam attack would take 3 hours then. During that time, apps based on WoT would be downloading the spam, rendering themselves useless for the user. - When I briefly read LCWoT's trust algorithm, it seemed to me like it doesn't compute the same thing as WoT. I'm not 100% sure there, but rather confident. So we'd be replacing a known-to-work algorithm with something rather different. - It doesn't yet ship the event-notifications API, which WoT client application maintainers should *URGENTLY* switch to. The existing API is a mess, as it *CANNOT* provide performance by design. I.e. it's not an implementation problem, the API just sucks to the point where it cannot be implemented in a fast way. In fact, I would say that it is very likely that WoT's performance is only half of the problem. The other half is the fact that existing clients use the old API; and do that in a very excessive way: Sone still downloads almost the whole WoT database every 60 seconds, where it rather should do this something like once every 60 *minutes*: https://github.com/Bombe/Sone/pull/11 So if you want to do something to improve the situation, I would recommend one of those: - Apply gentle annoyance to client app maintainers to ensure they fix their old-API delay to 60 minutes or above. - Help migrate client apps to use the new event-notifications API. See my flog post about build0015 for how to get started with it.
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Devl mailing list Devl@freenetproject.org https://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl
