Re: Is this the devl mailing list for Hyphanet/Freenet_Classic?
I hadn't thought of that, I assumed it is, but that's actually a good question... While we're talking about good questions and the allocation of old resources, what happens to the seemingly large number of bitcoins that were donated to former-Freenet-now-Hyphanet? Last time I heard of them, there was "plenty of" them (https://www.mail-archive.com/devl@freenetproject.org/msg55173.html), which is a bit vague, but looking at the former donation address (https://www.blockchain.com/explorer/addresses/btc/1966U1pjj15tLxPXZ19U48c99EJDkdXeqb if I'm not mistaken - https://www.mail-archive.com/devl@freenetproject.org/msg25689.html) this could be in the order of a few hundreds? Since the bulk of donations to this address was made before 2020 (or even before 2015, really), I'm doubtful that they were intended for former-Locutus-new-Freenet, which had its first commit apparently on Jul 16, 2021 (https://github.com/freenet/freenet-core/commit/c58ed0e86fa1a4f0389d85ded8c861ee6efa96c5). On Fri, Feb 2, 2024, at 21:07, Juiceman wrote: > Is this the devl mailing list for Hyphanet/Freenet_Classic? > -- > I may disagree with what you have to say, but I shall defend, to the > death, your right to say it. - Voltaire > Those who would give up Liberty, to purchase temporary Safety, deserve > neither Liberty nor Safety. - Ben Franklin
Re: Important Announcement: Freenet naming change
On Thu, Jan 19, 2023, at 00:13, Ian Clarke wrote: > On Wed, Jan 18, 2023, 4:47 PM David Dernoncourt > wrote: >> On Wed, Jan 18, 2023, at 19:32, Ian Clarke wrote: >> > That's not what's happening. The name always belonged to the *mission*, >> > not to any particular piece of software. This is why we were careful >> > not to tie it to any particular codebase when we wrote it: >> >> Maybe that's how you intended it originally, but with the possible >> exceptions of the devs who work on the depot named "fred" and maybe some >> users who are really experts and die-hard on wording accuracy, what most >> users call "Freenet" is that piece of software that you download from on >> this page https://freenetproject.org/pages/download.html that begins with >> "Download and install Freenet:" > > Firstly, you have no idea how most users of Fred feel about this > change, so I hope you're not implying that they all agree with you. > There is a lot of "argumentum ad populum" going on in this thread. > People should speak for themselves. Where did I mention agreement to the change in what you quote? All I'm saying is that the name "Freenet" is currently attached, in most people's minds, to the particular software and not to "the mission". And before you go back with a "you can't know what people think" argument: why the Hell would you assume a random user who downloads the only software offered on a page subtitled "download Freenet" (and named "FreenetInstaller.exe") would think "Freenet" isn't the name of the software they download and install? Until otherwise proven, they downloaded and ran a software, not a mission statement. > But more importantly, let's pretend that every single Fred user > disagreed with changing "Freenet" to "Freenet Classic". As I've already > stated - the constituency I'm concerned with is the next generation. I don't see how the naming in itself will be a game changer "for the next generation". "The next generation" will adopt any product it deems as "cool", even if it has a stupid name such as "TikTok" or "FTX". So as far as they're concerned, the name doesn't even matter. So yeah, you could rename as you say. You could also rename Locutus as Dicky-Doo-Dah. That's not really where the problem lies. Just like Freenet (current) didn't fail because of its name, Freenet (new) won't succeed thanks to of its name. However, what about the current generation? Is it gently (or not so gently) being told to sod off? I'd like to know, cause stopping a few servers here and there would mean fewer bills to pay. > So if you've got a compelling factual argument for why you think not > doing this will be better for the next generation, then I'm all ears. As I said, for "the next generation", naming isn't important. You could even try naming Locutus "Big Brother", apparently people like that, and younger generations in particular. And considering your naming theory, maybe it's just because the name is catchy.
Re: Important Announcement: Freenet naming change
On Wed, Jan 18, 2023, at 19:32, Ian Clarke wrote: > On Wed, Jan 18, 2023 at 11:08 AM David Dernoncourt > wrote: >> As "just a regular user" and node runner who just poked around a few PRs, >> the main issues I have with this are 1) "what the Hell is Locutus" and 2) >> wow that's confusing. >> >> Sure, 1) is just (mostly) a rhetorical question. But it seems quite an >> unusual thing to "steal" the name of a software (even more so when said >> software is still actively maintained) to stamp it on another software that >> doesn't have any relationship with the other one beside being supported by >> the same foundation and sharing the same (or a similar enough) mission >> statement. > > That's not what's happening. The name always belonged to the *mission*, > not to any particular piece of software. This is why we were careful > not to tie it to any particular codebase when we wrote it: > Maybe that's how you intended it originally, but with the possible exceptions of the devs who work on the depot named "fred" and maybe some users who are really experts and die-hard on wording accuracy, what most users call "Freenet" is that piece of software that you download from on this page https://freenetproject.org/pages/download.html that begins with "Download and install Freenet:"
Re: Important Announcement: Freenet naming change
As "just a regular user" and node runner who just poked around a few PRs, the main issues I have with this are 1) "what the Hell is Locutus" and 2) wow that's confusing. Sure, 1) is just (mostly) a rhetorical question. But it seems quite an unusual thing to "steal" the name of a software (even more so when said software is still actively maintained) to stamp it on another software that doesn't have any relationship with the other one beside being supported by the same foundation and sharing the same (or a similar enough) mission statement. While I do understand the reasoning of "I can't come up with a better name to describe what we're doing", boy, how confusing is that. 20-something year old software suddenly becomes totally different. Imagine Mozilla rebranding "Thunderbird" as "Firefox" and "Firefox" as "Firefox Classic" (and on top of that you've never head of Thunderbird), what a mess this would be. As far as suggestions go, even though (if it's not obvious) I don't support this change either (I would rather prefer, for instance, "Freenet Locutus" vs "Freenet Classic", dropping "Freenet [period]" instead of hijacking it), my suggestion would be to find a way to make it crystal clear that names were swapped and to point out clearly and concisely the key differences between the "old" Freenet and the "new" one. On Tue, Jan 17, 2023, at 15:04, Ian Clarke wrote: > Dear Freenet users and developers, > > I hope this email finds you well. I am writing to inform you of an > important change that the Freenet Project board voted on unanimously on > Friday. After much discussion over the past 18 months, we have decided > to rename the codebase and software known internally as "Fred," to > "Freenet Classic." > > In addition, we are announcing that the new codebase, known internally > as "Locutus," will be renamed to "Freenet." We understand that this is > a big change and it carries risk, but the board believes that this risk > is necessary in order to further the project's mission. > > It is important to note that this change does *not* mean that Locutus > is replacing Fred, which solves related but different problems. We will > ensure that where Fred is the appropriate tool for people they will be > directed to it. Freenetproject.org will remain focused on Fred while > linking to Locutus - as it has been for over two decades - while > freenet.org will place more emphasis on Locutus, while still linking to > Fred. > > We will be implementing these changes over the coming days and weeks > and would be grateful for your ideas and suggestions on how best to do > this. > > Thank you, > > Ian > > -- > Ian Clarke > Founder, The Freenet Project > Email: i...@freenetproject.org
Re: Please help doing the three more checks required for releasing 1493!
Hi, >> 1. [ ] auto update on windows I tested auto-update on one of my Windows nodes (with Java 11) and it successfully auto-updated to 1493. Another of my Windows nodes, running 1492 and on which I didn't touch the update key, received an update notice too: was the update released already? (this node is on manual update and I didn't apply it yet) On Mon, Apr 4, 2022, at 21:19, Dr. Arne Babenhauserheide wrote: > Hi, > > > after fixing a bug that turned up in testing (wrong CHK in > dependencies.properties that got aborted because it did not match the > correct sha256 — I’m glad we double-check!), we need *three more tests* > not done by me to be able to finally push 1493 to the website and to > auto-update (because you seriously don’t want me to be the only one who > tests this). > > > Please help get the release out! > > > Required tests: > >> 1. [ ] auto update on windows >> 2. [ ] auto update on GNU/Linux >> 3. [ ] installer on windows >> 4. [X] installer on GNU/Linux — thanks David! ✓ > > > Practically this means: > > 1. If you have a windows node, go to >http://127.0.0.1:/config/node.updater?fproxyAdvancedMode=2 >then put the new testing key into the field that says >"Where should Freenet look for updates?" > > USK@BrNh~RNzsl3zQueAH0Ed8bgF88kZHa4AH64RNKjsCU4,~hvYp2qtiUUXk4r2AuwMbiNvLiBcPhl9Nt4lsrvaYn8,AQACAAE/jar/1492 >Then restart your node. >It should update to 1493. Please report if you run into trouble! >Also please report if it works! > > 2. Do the same on GNU/Linux. Yes, I did it, but if I’m the only one who >does it, how should you trust Freenet to actually work? > > 3. Download the installer on Windows, run it, start the installed >Freenet, and report errors you see — or that it works. >The installer: > > CHK@07dc2TkhlpuQBJV2WEK1z2SAycEiv07zbtZC9A4keQs,34RYMOAnxxIkGAyLIfC8qNoSNuZpPRjWtvS68aZ5rbQ,AAMC--8/FreenetInstaller-1493.exe >or > > https://github.com/freenet/fred/releases/download/build01493/FreenetInstaller-1493.exe > > > Once these three checks are done (and worked), I can finally push the > update. > > > … and once I pushed it, please run verify-build to ensure that what I > released is actually what’s in the sources! > > - https://github.com/freenet/scripts/blob/master/verify-build > > > Let’s get this release out together! > > > Best wishes, > Arne > -- > Unpolitisch sein > heißt politisch sein, > ohne es zu merken. > draketo.de > > Attachments: > * signature.asc
Re: New test-release for 1493, with new testing key
I installed successfully on Linux with Java 17. Default datastore size was 100 GiB indeed. On a side note, it would be great to be able to define a custom value without having to live-edit the page with web dev tools, like it is possible for bandwidth settings. Default upload speed was 32 KiB/s. Default number of threads was 1000. All default bookmarks worked. Thanks for all the continued work :) On Sun, Apr 3, 2022, at 01:09, Dr. Arne Bab. wrote: > Thanks to TheSeeker for the review! > > > We found an error in dependencies.properties: wrapper-linux-ppc-64 had > the wrong CHK key. This should not have any effect on any other > platform, but better safe than sorry. > > Updated jars and installers: > https://github.com/freenet/fred/releases/tag/build01493 > > CHK@guwY9~yFAK-54jJFfvnCiSUShtHKDbWBdNXRRZmSeYc,N9kLHcWeKD1RzvuqEC8MvpDuBX6pH4PxwQlgc1P-5xM,AAMC--8/freenet-build01493.jar > CHK@IWmuNY6SuaWE-fn86YO~pqlDrVcDjADZ0T6Ew-ZL9MA,AKpxdi2Bm9BQu1Fs1hMoagn3vdeB1IGub~nX5BQMUhY,AAMC--8/freenet-build01493.jar.sig > CHK@07dc2TkhlpuQBJV2WEK1z2SAycEiv07zbtZC9A4keQs,34RYMOAnxxIkGAyLIfC8qNoSNuZpPRjWtvS68aZ5rbQ,AAMC--8/FreenetInstaller-1493.exe > CHK@4qXfy7qBJ-UGWJ51Jpfd2A5xlU0wPBVOR6dUiomrilU,uQh6xGyUJwlDZSec-LYTmuqf7g7p2e2fvXJuevq4o~E,AAMC--8/FreenetInstaller-1493.exe.sig > CHK@AufV75DZTW2wMIcQ5u0VsWXqiLguMOHK4jnpuhnmjU8,ZMJcjmuUlLO6ojpsGjjTTbbBjw~Q~TzmelbWxf0beEc,AAMC--8/new_installer_offline_1493.jar > CHK@nqIPBJMvPyocdir3Kky7bF3sBKDWj-DOFe4kmxfju-c,BMOAEJRVRhXx81B~-wd~h~geiS-HQrZmLe6rFM6ClHY,AAMC--8/new_installer_offline_1493.jar.sig > > CHK@Gxxk-ycyO~k8Cg62~9B1LHyZlzFoGouIuf5uE1THXrQ,31RDTUGUeJ8dXgAM8Q3ZAdlRJHvia9KbxrYRoYpEbVU,AAMC--8/freenet-build01493-source.tar.bz2 > > > > There’s also a new testing update key: > USK@BrNh~RNzsl3zQueAH0Ed8bgF88kZHa4AH64RNKjsCU4,~hvYp2qtiUUXk4r2AuwMbiNvLiBcPhl9Nt4lsrvaYn8,AQACAAE/jar/1492 > > The inserts to the testing key are running right now. Please give it a > go! Since this is a tiny change compared to the test release from last > weekend, I plan to release when I get one positive report for: > - auto update on windows > - auto update on GNU/Linux > - installer on windows > - installer on GNU/Linux > > (just to make sure nothing "theoretically impossible" broke). > > > Also please comment on the release notes; I plan to re-use these for the > announcement on the website: > > > -- Draft for Announcement -- > > Freenet 0.7.5 build 1493 is now available. [overview] > > > This build provides four core improvements: > > 1. Curated default bookmarks, >including an actively maintained index and Shoeshop for sneakernet > 2. Better peer scaling for very fast nodes > 3. Updated defaults to adapt to the higher capacities of modern systems > 4. Compatibility with Java 17, first in the installers, with the >following update for all nodes > > > The curated default bookmarks provide a much better first-start > experience. Previously new users saw mostly outdated sites in > inactive indexes. > > Adding Shoeshop to enable sneakernet which can connect separate > Freenet networks even if no internet connection can be established > now provides all the tools for selfpublishing, not only in the > style of publishing an online blog (which is already easy with > Sharesite) or sending a file to an independent printer, but in the > much more self-reliant style, resourcefulness and rebellious spirit > of classical samizdat. > > > Adjusted peer-scaling fixes a conceptual problem: Fast nodes could > not utilize their bandwidth well enough, because the previous > peer-scaling did not take the aggregated bandwidth limit of the > peers into account. Now very fast nodes have linearly scaling > peer-counts to make it more likely that the capacity of their peers > added together matches the capacity of the fast node. The absolute > upper limits stay in place, because they are needed to preserve > privacy. There are also no changes to the peer-scaling of slower > nodes. > > The fixed scaling should improve the performance of the whole > network because it avoids creating artificial bottlenecks. > > > Changes to the defaults are a doubled thread limit of newly > installed nodes (increased from 500 to 1000), with the stack size > per thread reduced by half to avoid higher memory consumption, the > datastore size is increased from 20GiB to 100GiB, because SSDs are > much faster and more resilient than before, and the default > bandwidth to offer if the actual speed cannot be found is doubled > to 32KiB/s. > > These newer defaults should also improve the first-time user > experience. > > > Compatibility with Java 17 took longer than we hoped, because it > required deploying a newer wrapper and changes to the classpath. > This makes it easier to support packages for modern Linux, and it > should avoid losing nodes when Java updates itself (starting from > the next update this also applies to existing nodes; we have to > deploy the update code in 1493 so it
