From nextgens at freenetproject.org Sun Jun 1 12:43:11 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Sun, 1 Jun 2008 14:43:11 +0200 Subject: [freenet-dev] Trust path, new download location Message-ID: <20080601124310.GB3543@freenetproject.org> Hi, Over the WE I have been changing a few things on emu... Three goals here: 1) Make the installation and update process MITM proof [1] 2) Get rid of the current load-balancer (I wrote it in php and it sucks...) and limit the number of redirects to follow to download a file 3) Gradually get rid of all the backward compatibility stuffs we keep server-side Hopefully that means less hits on emu and faster downloads. From now on all the files are downloaded from https://checksums.freenetproject.org/latest/ ... And no, don't ask for it there is no directory listing... We need to be able to change the URL we use more easily than now; Hence I've simplified to the minimum the naming scheme: you ask for the filename and that's it; they are no subdirectories anymore and latest version of plugins can be accessed directly. Before I can finish up and move on they are three things which should be done: 1) update.cmd needs to be updated; imho it could be rewritten from scratch, using update.sh as model. I can precisely describe why it sucks if needed. We want all the scripts to use the Sha1Test class 2) The node should be updated to download plugins from the new URL and should check the checksums it provides 3) users should be asked to double-check that they have an up to date updating script and that *it works*; It's probably something we want to talk about in the next release-announcement message. Once those are done I will get rid of two vhosts on emu (downloads.freenetproject.org and get.freenetproject.org) and simplify *a lot* the release-building scripts. I'll create a vhost dedicated to the 0.5 legacy code if someone want it (complain now or never)... but we can't reasonably keep the current file hierarchy. Would anyone object if I force the installer to be downloaded from emu and disable file-listing in the directory where it will be ? Lately I heard of weird bugs from some users... and it turns out they were using a 7 month old installer to set their nodes up... they used some download-directory (like zdnet.com) and got their old, deprecated installer from them! Needless to say there isn't much we can do against that... but checking referrers and disabling file-listing is something we can do reasonably easily. Are Zero3Cool or Juiceman still around and willing to tackle the update.cmd part or shall I add it to my todo? Does anyone want to do the plugins part? Bombe? dbkr? NextGen$ [1] Here is the security model: The certificate of checksums.freenetproject.org has been signed by a real CA... We bundle it in the installer and verify the certificates with it (It is valid 'till 2036); Arguably we should check some CRLs too... Files whose extensions are .sig and .sha1 are downloaded directly from emu, other filetypes lead to a direct redirection to one of the mirrors. We check the file we got from the mirror against the SHA1 checksum we downloaded securely from emu. Everything is handled by apache through RewriteRules, no php vodoo is involved anymore :) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080601/fb5e13ad/attachment.pgp From ian.clarke at gmail.com Mon Jun 2 02:12:07 2008 From: ian.clarke at gmail.com (Ian Clarke) Date: Sun, 1 Jun 2008 21:12:07 -0500 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on Windows XP In-Reply-To: <200805301757.49729.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> Message-ID: <823242bd0806011912qc4c3ec8r207b91e6500878d9@mail.gmail.com> Should probably get these into the bugtracker. On 5/30/08, Matthew Toseland wrote: > > > ---------- Forwarded Message ---------- > > Subject: [freenet-support] Usability testing on Windows XP > Date: Thursday 29 May 2008 15:52 > From: Julien Cornuwel > To: support at freenetproject.org > > Hi, > > > Just had the occasion to follow a beginner through the installation > process via Java Webstart, on Windows XP. Here is what I noticed : > > - The installer is now signed with a Thawte certificate, which is > accepted by default. That's good ! > - The two phase install confused the user. He thought the install was > finished after the first one and didn't understood the programm was > actually downloading freenet in the second one. > - He was again confused when the browser spawned up saying "Installation > successfull" and the installer was still donwloading stuff. > Additionnaly, that page was in English despite the fact he chose French > in the installer. I had to tell him he had to finish the installer and > then, click on the link to the first-time wizard. > - The wizard was quite straightforward and he choose default values, > which happened to be good for him, IMHO. > - Just a detail : every shortcuts (except Browse Freenet) in the menu > came with the same default icon. He was disapointed and I had to tell > him it was irrelevant and everything was fine. > > Maybe we should only spawn the browser at the very end of the > installation process, when the installer closes and Freenet is up and > running. And maybe we should open directly to the wizard page, which is > i18n-ed. > > > Hope this is usefull... > > > > ------------------------------------------------------- > -- Sent from Gmail for mobile | mobile.google.com Email: ian at uprizer.com Cell: +1 512 422 3588 Skype: sanity From toad at amphibian.dyndns.org Mon Jun 2 12:57:15 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Mon, 2 Jun 2008 13:57:15 +0100 Subject: [freenet-dev] Custom Firefox profile causing aggravation with users? In-Reply-To: <48414FEA.4030107@users.sourceforge.net> References: <48360CD9.9080307@redline.ru> <200805271616.45318.toad@amphibian.dyndns.org> <48414FEA.4030107@users.sourceforge.net> Message-ID: <200806021357.22896.toad@amphibian.dyndns.org> On Saturday 31 May 2008 14:17, Michael T?nzer wrote: > Matthew Toseland schrieb: > > On Sunday 25 May 2008 18:42, Michael T?nzer wrote: > >> Matthew Toseland schrieb: > >>> On Friday 23 May 2008 22:40, Michael T?nzer wrote: > >>>> Matthew Toseland schrieb: > >>>>> On Friday 23 May 2008 11:38, Matthew Toseland wrote: > >>>>>>> I'm not pushing for any immediate changes, but perhaps being more > >>>>>>> user-friendly regarding the custom FF profile is something to consider > >>>>>>> for 0.7.1? > >>>>>> I'd welcome any suggestions. So far, afaics the options are: > >>>>>> 1) Fix the Firefox bug that causes the profile resetting. -no-remote > >>> should > >>>>>> cause it not only to not coalesce with an existing Firefox copy, but > > also > >>>>> not > >>>>>> to write to the default profile. Also find a new skin that works with > >>> FF3, > >>>>>> and ideally is a little more stable! > >>>>>> 2) Build something using XULRunner. I believe this is the recommended > > way > >>> of > >>>>>> doing things according to the Firefox devs. They provide a sample > > browser > >>>>>> implemented in XUL, but it's *really* minimal, no right-click-go-back > > for > >>>>>> example. > >>>>>> 3) Bundle a "portable" browser. > >>>>> 4) Bundle a browser plugin. > >>>> Maybe we can steal some code from this plugin: > >>>> https://nic-nac-project.org/~kaosmos/profileswitcher-en.html > >>>> > >>>> it's under GPLv3 (source is in the .xpi) and as far as I have tested, it > >>>> doesn't have the bug you described. So we could still provide a > >>>> Firefox-Profile (which isn't used by default) but also ask (in the > >>>> install wizard) to install the plugin. We alter it's behaviour in such a > >>>> way, that it simply provides a "Browse Freenet" button or menu item > >>>> which then loads our custom profile. > >>>> The advantage would be, that the user has to opt-in and that while he's > >>>> in the browser he can easily browse Freenet (no need to switch through > >>>> start menus). > >>> Don't we want the user to be able to browse Freenet (in a customised > > browser) > >>> and non-freenet sites at the same time? > >>> > >> Yes we would still have the two profiles method, the user has two > >> windows open, one with the Freenet profile and one with his default one. > >> We probably don't want Freenet and usual webpages in one window, because > >> it leads to having to use the same profile (I don't think you can do two > >> profiles in one window) which would affect both, performance and privacy. > >> The advantage of the button method is, that the user knows there where > >> adjustments done to his browser (because he had to confirm the install > >> of the plugin in Firefox), we provide a shortcut to the Freenet profile > >> (no hangling through menus) and it doesn't have the bug, that Firefox > >> uses the Freenet profile as the default if you close the default before > >> you close the Freenet profile (at least I haven't experienced it up > >> until now). > > > > So you can switch profiles on a per-window basis? > > > > Well, if you select another profile in the menu (or in our case click on > the "Browse Freenet" button) you will be asked whether you want to close > the current profile/window (you can also set a default in the settings > and you will not be asked, in our case setting the default to not > closing the current window might be reasonable) and then another window > with the new profile comes up. > So you can't switch the profile of the current window but you can open > another one with the selected profile. > Well, if it doesn't change the default profile, this might be an option... -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/7e1694c3/attachment.pgp From toad at amphibian.dyndns.org Mon Jun 2 13:15:02 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Mon, 2 Jun 2008 14:15:02 +0100 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on Windows XP In-Reply-To: <200805301757.49729.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> Message-ID: <200806021415.07840.toad@amphibian.dyndns.org> On Friday 30 May 2008 17:57, Matthew Toseland wrote: > > ---------- Forwarded Message ---------- > > Subject: [freenet-support] Usability testing on Windows XP > Date: Thursday 29 May 2008 15:52 > From: Julien Cornuwel > To: support at freenetproject.org > > Hi, > > > Just had the occasion to follow a beginner through the installation > process via Java Webstart, on Windows XP. Here is what I noticed : > > - The installer is now signed with a Thawte certificate, which is > accepted by default. That's good ! > - The two phase install confused the user. He thought the install was > finished after the first one and didn't understood the programm was > actually downloading freenet in the second one. I assume by two phase install you mean the copying files page and the processes page? I'm not sure there's much we can do about this, except perhaps to make the processes page take less time. > - He was again confused when the browser spawned up saying "Installation > successfull" and the installer was still donwloading stuff. This is a compromize because we have some largish stuff that needs to be downloaded after starting the node: Thaw, Thingamablog, etc. Arguably the solution is to debundle them. > Additionnaly, that page was in English despite the fact he chose French > in the installer. I had to tell him he had to finish the installer and > then, click on the link to the first-time wizard. Okay this is bad, I think it's been fixed though. Nextgens? > - The wizard was quite straightforward and he choose default values, > which happened to be good for him, IMHO. The default values should be acceptable for most people, that's why they're default values. > - Just a detail : every shortcuts (except Browse Freenet) in the menu > came with the same default icon. He was disapointed and I had to tell > him it was irrelevant and everything was fine. > > Maybe we should only spawn the browser at the very end of the > installation process, when the installer closes and Freenet is up and > running. We open the browser very shortly after Freenet is working. However, at the moment, we download and install more stuff after that - namely the bundled apps. In any case, without major changes to the installer, we have to open the browser during the processes page (again, am I right nextgens? Is there any possibility of exit-hooks?). Is it a big deal? The most likely consequence is that the user doesn't complete the installer, or that there is a long delay before it, right? > And maybe we should open directly to the wizard page, which is > i18n-ed. The welcome page is l10n'ed, however the question is how much benefit do we gain by opening it instead of the config wizard? > > Hope this is usefull... Okay, what issues do we have here? 1. The processes page takes longer than the copying files page. In fact, most of the time of the installer is occupied by the processes page. IMHO the solution to this is to bundle everything which is essential with the base installer. Then the processes page will take much less time. Also we won't need a separate bundle-installer, so it will be easier to use in places where the Freenet website is blocked. The downside is the installer will be significantly bigger. 2. The installer shouldn't download Thingamablog and Thaw. They are huge, we have not audited the code, and users can get them if they want them. Debundling them would mean that as soon as we open the browser the Processes page will have completed, and the user can click Next and go on to creating icons. There seems to be a consensus on debundling, at least between Ian and Nextgens. :) I think a policy of not bundling anything we can't code review is reasonable, and clearly code reviewing Thingamablog for example would be a massive undertaking and isn't appropriate at this time. I do think we should keep the current plugins, however most of them are small. Even if we include the WoT plugin, that is also likely to be reasonably small for the foreseeable future. 3. Is it possible to change the names of the two panels? Copying Files vs Setting Up Freenet, perhaps? 4. Is it possible to open the browser after closing the installer rather than in the Processes page? 5. Icons: We need proper icons for our various shortcuts. On unix the situation is worse: We have no icons. On Windows, Browse Freenet has the bunny icon and the rest don't have anything. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/0e427b99/attachment.pgp From nextgens at freenetproject.org Mon Jun 2 13:43:59 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Mon, 2 Jun 2008 15:43:59 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on Windows XP In-Reply-To: <200806021415.07840.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021415.07840.toad@amphibian.dyndns.org> Message-ID: <20080602134357.GD3513@freenetproject.org> * Matthew Toseland [2008-06-02 14:15:02]: > > Additionnaly, that page was in English despite the fact he chose French > > in the installer. I had to tell him he had to finish the installer and > > then, click on the link to the first-time wizard. > > Okay this is bad, I think it's been fixed though. Nextgens? > Yeah, it was a windows-specific bug and which has been fixed. > > - Just a detail : every shortcuts (except Browse Freenet) in the menu > > came with the same default icon. He was disapointed and I had to tell > > him it was irrelevant and everything was fine. > > If you have icon files I can use I'll be pleased to fix that > > Maybe we should only spawn the browser at the very end of the > > installation process, when the installer closes and Freenet is up and > > running. > > We open the browser very shortly after Freenet is working. However, at the > moment, we download and install more stuff after that - namely the bundled > apps. In any case, without major changes to the installer, we have to open > the browser during the processes page (again, am I right nextgens? Is there > any possibility of exit-hooks?). Is it a big deal? Last time I tried to use them they were issues > The most likely consequence is that the user doesn't complete the installer, > or that there is a long delay before it, right? > We could time-delay the browser startup if that's the only thing :) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/6202c459/attachment.pgp From nextgens at freenetproject.org Mon Jun 2 13:47:20 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Mon, 2 Jun 2008 15:47:20 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on Windows XP In-Reply-To: <200806021415.07840.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021415.07840.toad@amphibian.dyndns.org> Message-ID: <20080602134718.GE3513@freenetproject.org> * Matthew Toseland [2008-06-02 14:15:02]: > Okay, what issues do we have here? > > 1. The processes page takes longer than the copying files page. In fact, most > of the time of the installer is occupied by the processes page. > > IMHO the solution to this is to bundle everything which is essential with the > base installer. Then the processes page will take much less time. Also we > won't need a separate bundle-installer, so it will be easier to use in places > where the Freenet website is blocked. The downside is the installer will be > significantly bigger. > I'm not convinced about that > 2. The installer shouldn't download Thingamablog and Thaw. They are huge, we > have not audited the code, and users can get them if they want them. > Debundling them would mean that as soon as we open the browser the Processes > page will have completed, and the user can click Next and go on to creating > icons. There seems to be a consensus on debundling, at least between Ian and > Nextgens. :) I think a policy of not bundling anything we can't code review > is reasonable, and clearly code reviewing Thingamablog for example would be a > massive undertaking and isn't appropriate at this time. I do think we should > keep the current plugins, however most of them are small. Even if we include > the WoT plugin, that is also likely to be reasonably small for the > foreseeable future. > Ok, let's get rid of them then > 3. Is it possible to change the names of the two panels? Copying Files vs > Setting Up Freenet, perhaps? > We could get rid of one of the panel before the processing one > 4. Is it possible to open the browser after closing the installer rather than > in the Processes page? > Not "cleanly" but we could do a workaround time-delaying the browser startup. > 5. Icons: We need proper icons for our various shortcuts. On unix the > situation is worse: We have no icons. On Windows, Browse Freenet has the > bunny icon and the rest don't have anything. Well provide me icons and maybe I'll consider using them :p -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/0aaa4b25/attachment.pgp From toad at amphibian.dyndns.org Mon Jun 2 14:03:37 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Mon, 2 Jun 2008 15:03:37 +0100 Subject: [freenet-dev] =?utf-8?q?Fwd=3A_=5Bfreenet-support=5D_Usability_te?= =?utf-8?q?sting_on=09Windows_XP?= In-Reply-To: <20080602134718.GE3513@freenetproject.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021415.07840.toad@amphibian.dyndns.org> <20080602134718.GE3513@freenetproject.org> Message-ID: <200806021503.42797.toad@amphibian.dyndns.org> On Monday 02 June 2008 14:47, Florent Daigni?re wrote: > * Matthew Toseland [2008-06-02 14:15:02]: > > > Okay, what issues do we have here? > > > > 1. The processes page takes longer than the copying files page. In fact, most > > of the time of the installer is occupied by the processes page. > > > > IMHO the solution to this is to bundle everything which is essential with the > > base installer. Then the processes page will take much less time. Also we > > won't need a separate bundle-installer, so it will be easier to use in places > > where the Freenet website is blocked. The downside is the installer will be > > significantly bigger. > > I'm not convinced about that Why? It won't be huge, if we don't bundle any unnecessary Huge Stuff (see below). And it's not small now. > > > 2. The installer shouldn't download Thingamablog and Thaw. They are huge, we > > have not audited the code, and users can get them if they want them. > > Debundling them would mean that as soon as we open the browser the Processes > > page will have completed, and the user can click Next and go on to creating > > icons. There seems to be a consensus on debundling, at least between Ian and > > Nextgens. :) I think a policy of not bundling anything we can't code review > > is reasonable, and clearly code reviewing Thingamablog for example would be a > > massive undertaking and isn't appropriate at this time. I do think we should > > keep the current plugins, however most of them are small. Even if we include > > the WoT plugin, that is also likely to be reasonably small for the > > foreseeable future. > > Ok, let's get rid of them then For reasons of code reviewability? Further to the above, I do actually review Thingamablog commits, however I haven't reviewed the original, and it's not really widely enough used that we can assume it to be safe? However, I do disagree with Ian's assertion that nothing we don't own is reviewable: it is much easier to hide bugs in C because you have buffer overflows, format string vulnerabilities and so on. In Java, well written code *can* be reviewed, although of course more subtle bugs may slip through the net. Also I review jSite commits (I dunno whether the original on which the diffs are applied was reviewed though). > > > 3. Is it possible to change the names of the two panels? Copying Files vs > > Setting Up Freenet, perhaps? > > We could get rid of one of the panel before the processing one Good idea. > > > 4. Is it possible to open the browser after closing the installer rather than > > in the Processes page? > > Not "cleanly" but we could do a workaround time-delaying the browser > startup. I'm not convinced that's better than the alternative. > > > 5. Icons: We need proper icons for our various shortcuts. On unix the > > situation is worse: We have no icons. On Windows, Browse Freenet has the > > bunny icon and the rest don't have anything. > > Well provide me icons and maybe I'll consider using them :p https://bugs.freenetproject.org/view.php?id=2409 https://bugs.freenetproject.org/view.php?id=2026 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/d517c041/attachment.pgp From nextgens at freenetproject.org Mon Jun 2 14:13:13 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Mon, 2 Jun 2008 16:13:13 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on?Windows XP In-Reply-To: <200806021503.42797.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021415.07840.toad@amphibian.dyndns.org> <20080602134718.GE3513@freenetproject.org> <200806021503.42797.toad@amphibian.dyndns.org> Message-ID: <20080602141312.GF3513@freenetproject.org> * Matthew Toseland [2008-06-02 15:03:37]: > On Monday 02 June 2008 14:47, Florent Daigni?re wrote: > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > Okay, what issues do we have here? > > > > > > 1. The processes page takes longer than the copying files page. In fact, > most > > > of the time of the installer is occupied by the processes page. > > > > > > IMHO the solution to this is to bundle everything which is essential with > the > > > base installer. Then the processes page will take much less time. Also we > > > won't need a separate bundle-installer, so it will be easier to use in > places > > > where the Freenet website is blocked. The downside is the installer will > be > > > significantly bigger. > > > > I'm not convinced about that > > Why? It won't be huge, if we don't bundle any unnecessary Huge Stuff (see > below). And it's not small now. That means we will have to generate new installers on a regular basis... I don't like that :p > > > 2. The installer shouldn't download Thingamablog and Thaw. They are huge, > we > > > have not audited the code, and users can get them if they want them. > > > Debundling them would mean that as soon as we open the browser the > Processes > > > page will have completed, and the user can click Next and go on to > creating > > > icons. There seems to be a consensus on debundling, at least between Ian > and > > > Nextgens. :) I think a policy of not bundling anything we can't code > review > > > is reasonable, and clearly code reviewing Thingamablog for example would > be a > > > massive undertaking and isn't appropriate at this time. I do think we > should > > > keep the current plugins, however most of them are small. Even if we > include > > > the WoT plugin, that is also likely to be reasonably small for the > > > foreseeable future. > > > > Ok, let's get rid of them then > > For reasons of code reviewability? > Yep > Further to the above, I do actually review Thingamablog commits, however I > haven't reviewed the original, and it's not really widely enough used that we > can assume it to be safe? However, I do disagree with Ian's assertion that > nothing we don't own is reviewable: it is much easier to hide bugs in C > because you have buffer overflows, format string vulnerabilities and so on. > In Java, well written code *can* be reviewed, although of course more subtle > bugs may slip through the net. Also I review jSite commits (I dunno whether > the original on which the diffs are applied was reviewed though). > > > > > 3. Is it possible to change the names of the two panels? Copying Files vs > > > Setting Up Freenet, perhaps? > > > > We could get rid of one of the panel before the processing one > > Good idea. Will see what I can do > > > > > 4. Is it possible to open the browser after closing the installer rather > than > > > in the Processes page? > > > > Not "cleanly" but we could do a workaround time-delaying the browser > > startup. > > I'm not convinced that's better than the alternative. Okay, so WONTFIX -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/8b9041c3/attachment.pgp From batosai at batosai.net Mon Jun 2 16:37:59 2008 From: batosai at batosai.net (Julien Cornuwel) Date: Mon, 02 Jun 2008 18:37:59 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on Windows XP In-Reply-To: <20080602134357.GD3513@freenetproject.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021415.07840.toad@amphibian.dyndns.org> <20080602134357.GD3513@freenetproject.org> Message-ID: <484421E7.7040708@batosai.net> Florent Daigni?re a ?crit : > * Matthew Toseland [2008-06-02 14:15:02]: > >>> Additionnaly, that page was in English despite the fact he chose French >>> in the installer. I had to tell him he had to finish the installer and >>> then, click on the link to the first-time wizard. >> Okay this is bad, I think it's been fixed though. Nextgens? >> > > Yeah, it was a windows-specific bug and which has been fixed. Before or after I did the test ? Because it was not at that moment. >>> - Just a detail : every shortcuts (except Browse Freenet) in the menu >>> came with the same default icon. He was disapointed and I had to tell >>> him it was irrelevant and everything was fine. >>> > > If you have icon files I can use I'll be pleased to fix that Well, no I don't *have* any. And seriously, you don't want me to draw some ;) >>> Maybe we should only spawn the browser at the very end of the >>> installation process, when the installer closes and Freenet is up and >>> running. >> We open the browser very shortly after Freenet is working. However, at the >> moment, we download and install more stuff after that - namely the bundled >> apps. In any case, without major changes to the installer, we have to open >> the browser during the processes page (again, am I right nextgens? Is there >> any possibility of exit-hooks?). Is it a big deal? > > Last time I tried to use them they were issues > >> The most likely consequence is that the user doesn't complete the installer, >> or that there is a long delay before it, right? >> > > We could time-delay the browser startup if that's the only thing :) Any solution would be good. Small details like that keep users telling others that Freenet is complicated. We are so used to it that we can't see them anymore but, newbies do... -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 252 bytes Desc: OpenPGP digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/64f91c50/attachment.pgp From batosai at batosai.net Mon Jun 2 16:49:10 2008 From: batosai at batosai.net (Julien Cornuwel) Date: Mon, 02 Jun 2008 18:49:10 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing on?Windows XP In-Reply-To: <20080602141312.GF3513@freenetproject.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021415.07840.toad@amphibian.dyndns.org> <20080602134718.GE3513@freenetproject.org> <200806021503.42797.toad@amphibian.dyndns.org> <20080602141312.GF3513@freenetproject.org> Message-ID: <48442486.2040004@batosai.net> Florent Daigni?re a ?crit : >>>> 4. Is it possible to open the browser after closing the installer rather >> than >>>> in the Processes page? >>> Not "cleanly" but we could do a workaround time-delaying the browser >>> startup. >> I'm not convinced that's better than the alternative. > > Okay, so WONTFIX What about simply not spawning the browser ? Users usually don't expect a program to be lauched right after installing, unless they asked for it. Once the installer finished, the user has a shiny shortcut on his desktop saying "Browse Freenet". There are good chances he will clic on it ;) Of course, there is the wizard : maybe the node should be able to detect that it isn't configured yet and redirect to the wizard page ? -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 252 bytes Desc: OpenPGP digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080602/005ada77/attachment.pgp From toad at amphibian.dyndns.org Tue Jun 3 16:22:56 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Tue, 3 Jun 2008 17:22:56 +0100 Subject: [freenet-dev] =?utf-8?q?Fwd=3A_=5Bfreenet-support=5D_Usability_te?= =?utf-8?q?sting=09on=09Windows_XP?= In-Reply-To: <484421E7.7040708@batosai.net> References: <200805301757.49729.toad@amphibian.dyndns.org> <20080602134357.GD3513@freenetproject.org> <484421E7.7040708@batosai.net> Message-ID: <200806031722.56480.toad@amphibian.dyndns.org> On Monday 02 June 2008 17:37, Julien Cornuwel wrote: > Florent Daigni?re a ?crit : > > * Matthew Toseland [2008-06-02 14:15:02]: > > > >>> Additionnaly, that page was in English despite the fact he chose French > >>> in the installer. I had to tell him he had to finish the installer and > >>> then, click on the link to the first-time wizard. > >> Okay this is bad, I think it's been fixed though. Nextgens? > >> > > > > Yeah, it was a windows-specific bug and which has been fixed. > > Before or after I did the test ? Because it was not at that moment. > > >>> - Just a detail : every shortcuts (except Browse Freenet) in the menu > >>> came with the same default icon. He was disapointed and I had to tell > >>> him it was irrelevant and everything was fine. > >>> > > > > If you have icon files I can use I'll be pleased to fix that > > Well, no I don't *have* any. And seriously, you don't want me to draw > some ;) > > >>> Maybe we should only spawn the browser at the very end of the > >>> installation process, when the installer closes and Freenet is up and > >>> running. > >> We open the browser very shortly after Freenet is working. However, at the > >> moment, we download and install more stuff after that - namely the bundled > >> apps. In any case, without major changes to the installer, we have to open > >> the browser during the processes page (again, am I right nextgens? Is there > >> any possibility of exit-hooks?). Is it a big deal? > > > > Last time I tried to use them they were issues > > > >> The most likely consequence is that the user doesn't complete the installer, > >> or that there is a long delay before it, right? > >> > > > > We could time-delay the browser startup if that's the only thing :) > > Any solution would be good. Small details like that keep users telling > others that Freenet is complicated. We are so used to it that we can't > see them anymore but, newbies do... I'm not sure that a time delay would help. And exit hooks don't work. So what can we do? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080603/28ce2eb2/attachment.pgp From toad at amphibian.dyndns.org Tue Jun 3 16:24:48 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Tue, 3 Jun 2008 17:24:48 +0100 Subject: [freenet-dev] =?utf-8?q?Fwd=3A_=5Bfreenet-support=5D_Usability_te?= =?utf-8?q?sting=09on=3FWindows_XP?= In-Reply-To: <20080602141312.GF3513@freenetproject.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021503.42797.toad@amphibian.dyndns.org> <20080602141312.GF3513@freenetproject.org> Message-ID: <200806031724.48961.toad@amphibian.dyndns.org> On Monday 02 June 2008 15:13, Florent Daigni?re wrote: > * Matthew Toseland [2008-06-02 15:03:37]: > > > On Monday 02 June 2008 14:47, Florent Daigni?re wrote: > > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > > > Okay, what issues do we have here? > > > > > > > > 1. The processes page takes longer than the copying files page. In fact, > > most > > > > of the time of the installer is occupied by the processes page. > > > > > > > > IMHO the solution to this is to bundle everything which is essential with > > the > > > > base installer. Then the processes page will take much less time. Also we > > > > won't need a separate bundle-installer, so it will be easier to use in > > places > > > > where the Freenet website is blocked. The downside is the installer will > > be > > > > significantly bigger. > > > > > > I'm not convinced about that > > > > Why? It won't be huge, if we don't bundle any unnecessary Huge Stuff (see > > below). And it's not small now. > > That means we will have to generate new installers on a regular basis... > I don't like that :p Is there any good reason for this dislike? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080603/c5640c6a/attachment.pgp From nextgens at freenetproject.org Wed Jun 4 04:35:52 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 06:35:52 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing?on?Windows XP In-Reply-To: <200806031722.56480.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <20080602134357.GD3513@freenetproject.org> <484421E7.7040708@batosai.net> <200806031722.56480.toad@amphibian.dyndns.org> Message-ID: <20080604043550.GD3506@freenetproject.org> * Matthew Toseland [2008-06-03 17:22:56]: > On Monday 02 June 2008 17:37, Julien Cornuwel wrote: > > Florent Daigni?re a ?crit : > > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > >>> Additionnaly, that page was in English despite the fact he chose French > > >>> in the installer. I had to tell him he had to finish the installer and > > >>> then, click on the link to the first-time wizard. > > >> Okay this is bad, I think it's been fixed though. Nextgens? > > >> > > > > > > Yeah, it was a windows-specific bug and which has been fixed. > > > > Before or after I did the test ? Because it was not at that moment. > > > > >>> - Just a detail : every shortcuts (except Browse Freenet) in the menu > > >>> came with the same default icon. He was disapointed and I had to tell > > >>> him it was irrelevant and everything was fine. > > >>> > > > > > > If you have icon files I can use I'll be pleased to fix that > > > > Well, no I don't *have* any. And seriously, you don't want me to draw > > some ;) > > > > >>> Maybe we should only spawn the browser at the very end of the > > >>> installation process, when the installer closes and Freenet is up and > > >>> running. > > >> We open the browser very shortly after Freenet is working. However, at > the > > >> moment, we download and install more stuff after that - namely the > bundled > > >> apps. In any case, without major changes to the installer, we have to > open > > >> the browser during the processes page (again, am I right nextgens? Is > there > > >> any possibility of exit-hooks?). Is it a big deal? > > > > > > Last time I tried to use them they were issues > > > > > >> The most likely consequence is that the user doesn't complete the > installer, > > >> or that there is a long delay before it, right? > > >> > > > > > > We could time-delay the browser startup if that's the only thing :) > > > > Any solution would be good. Small details like that keep users telling > > others that Freenet is complicated. We are so used to it that we can't > > see them anymore but, newbies do... > > I'm not sure that a time delay would help. And exit hooks don't work. So what > can we do? Maybe we should not open the browser and wait for the user to click on "browse-freenet". -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/49893624/attachment.pgp From nextgens at freenetproject.org Wed Jun 4 04:42:17 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 06:42:17 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing?on?Windows XP In-Reply-To: <200806031724.48961.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806021503.42797.toad@amphibian.dyndns.org> <20080602141312.GF3513@freenetproject.org> <200806031724.48961.toad@amphibian.dyndns.org> Message-ID: <20080604044216.GE3506@freenetproject.org> * Matthew Toseland [2008-06-03 17:24:48]: > On Monday 02 June 2008 15:13, Florent Daigni?re wrote: > > * Matthew Toseland [2008-06-02 15:03:37]: > > > > > On Monday 02 June 2008 14:47, Florent Daigni?re wrote: > > > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > > > > > Okay, what issues do we have here? > > > > > > > > > > 1. The processes page takes longer than the copying files page. In > fact, > > > most > > > > > of the time of the installer is occupied by the processes page. > > > > > > > > > > IMHO the solution to this is to bundle everything which is essential > with > > > the > > > > > base installer. Then the processes page will take much less time. Also > we > > > > > won't need a separate bundle-installer, so it will be easier to use in > > > places > > > > > where the Freenet website is blocked. The downside is the installer > will > > > be > > > > > significantly bigger. > > > > > > > > I'm not convinced about that > > > > > > Why? It won't be huge, if we don't bundle any unnecessary Huge Stuff (see > > > below). And it's not small now. > > > > That means we will have to generate new installers on a regular basis... > > I don't like that :p > > Is there any good reason for this dislike? Downloading stuffs during the install process actually tests the user's connectivity; If we don't do that we will end up with more support requests from users having non-freenet-related issues. Moreover, if we want the "trust-chain" in between us and the newbies to be established, we HAVE TO distribute the installer from emu and not from mirrors... meaning that it has to be as small as possible if we intend to stand a slashdoting. Arguably we don't distribute the installer ourselves yet... but I was in the process of changing that. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/d15efea0/attachment.pgp From nextgens at freenetproject.org Wed Jun 4 04:58:12 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 06:58:12 +0200 Subject: [freenet-dev] Website and IE Message-ID: <20080604045810.GG3506@freenetproject.org> The website looks awful in IE; is there any CSS wizard who has any clue on how to fix it ? http://img258.imageshack.us/img258/1830/72265808xc0.png -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/beb95329/attachment.pgp From nextgens at freenetproject.org Wed Jun 4 07:14:41 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 09:14:41 +0200 Subject: [freenet-dev] CPUID patch from FMS In-Reply-To: <20080518040914.GD3521@freenetproject.org> References: <200805171229.16815.toad@amphibian.dyndns.org> <20080517120308.GE3482@freenetproject.org> <20080517125841.GF3482@freenetproject.org> <200805171444.34705.toad@amphibian.dyndns.org> <20080518040914.GD3521@freenetproject.org> Message-ID: <20080604071440.GH3506@freenetproject.org> * Florent Daigni?re [2008-05-18 06:09:15]: > * Matthew Toseland [2008-05-17 14:44:27]: > > > On Saturday 17 May 2008 13:58, Florent Daigni?re wrote: > > > * Florent Daigni?re [2008-05-17 14:03:10]: > > > > > > > * Matthew Toseland [2008-05-17 12:29:09]: > > > > > > > > > As nextgens is worried about anonymous contributions, I thought he might > > want > > > > > to review this before I apply it. > > > > > > > > > > > > > There is a related bug in the bug tracker... It's not the only bits that > > > > need updating :/ > > > > > > > > See #74 > > > > > > http://download.intel.com/design/processor/applnots/24161832.pdf > > > > > http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/25481.pdf > > > > > > Here are the CPUID codes if you want to update them. > > > > > So should we or shouldn't we apply yongjhen's patch? > > I think that applying that patch instead of updating the whole set of codes > is not the way to go... > > If we were really after performances we would recompile the native > libraries with the new libgmp... which has performances boost on that > kind of (recent) processors. Any progress ? Did you fill a ticket in for that issue ? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/5f540378/attachment.pgp From Lokivgap at gmx.de Wed Jun 4 08:13:51 2008 From: Lokivgap at gmx.de (Loki) Date: Wed, 4 Jun 2008 10:13:51 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <20080604045810.GG3506@freenetproject.org> References: <20080604045810.GG3506@freenetproject.org> Message-ID: <20080604101351.ea19e51a.Lokivgap@gmx.de> Well not a wizard here, but maybe a help. In style.css just activate the out commented line /*width: 150px;*/ div#menu { font-family: Sans-serif, Arial, "Times New Roman", Georgia, Times, serif; float:left; /*width: 150px;*/ <---- font-style: normal; font-weight: normal; text-align: left; line-height: 100%; color: #000000; background: #CCDDFF; text-decoration: none; margin-top: 15px; margin-bottom: 10px; border: 1px solid #99CCFF; } Background: If no width definition for the Box exist, so Firefox use the minimum needed width, IE6 the maximum available width. Problems: If the box has a fixed width definition, it will not resize with the font zooming feature. On Wed, 4 Jun 2008 06:58:12 +0200 Florent Daigni?re wrote: > The website looks awful in IE; is there any CSS wizard who has any clue > on how to fix it ? > > http://img258.imageshack.us/img258/1830/72265808xc0.png > From nextgens at freenetproject.org Wed Jun 4 08:40:30 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 10:40:30 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <20080604101351.ea19e51a.Lokivgap@gmx.de> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> Message-ID: <20080604084028.GI3506@freenetproject.org> * Loki [2008-06-04 10:13:51]: > Well not a wizard here, but maybe a help. > > In style.css just activate the out commented line /*width: 150px;*/ > > div#menu { > font-family: Sans-serif, Arial, "Times New Roman", Georgia, Times, serif; > float:left; > /*width: 150px;*/ <---- > font-style: normal; > font-weight: normal; > text-align: left; > line-height: 100%; > color: #000000; > background: #CCDDFF; > text-decoration: none; > margin-top: 15px; > margin-bottom: 10px; > border: 1px solid #99CCFF; > } > > Background: > If no width definition for the Box exist, so Firefox use the minimum needed width, IE6 the maximum available width. > Okay, I've reverted r19832 in r20191 :) > Problems: > If the box has a fixed width definition, it will not resize with the font zooming feature. > Well, I guess we will have to cope with that for the time being :/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/18cd0314/attachment.pgp From NEOatNHNG at users.sourceforge.net Wed Jun 4 09:21:56 2008 From: NEOatNHNG at users.sourceforge.net (=?UTF-8?B?TWljaGFlbCBUw6RuemVy?=) Date: Wed, 04 Jun 2008 11:21:56 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <20080604084028.GI3506@freenetproject.org> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> <20080604084028.GI3506@freenetproject.org> Message-ID: <48465EB4.9040005@users.sourceforge.net> Florent Daigni?re schrieb: > * Loki [2008-06-04 10:13:51]: > >> Well not a wizard here, but maybe a help. >> >> In style.css just activate the out commented line /*width: 150px;*/ >> >> div#menu { >> font-family: Sans-serif, Arial, "Times New Roman", Georgia, Times, serif; >> float:left; >> /*width: 150px;*/ <---- >> font-style: normal; >> font-weight: normal; >> text-align: left; >> line-height: 100%; >> color: #000000; >> background: #CCDDFF; >> text-decoration: none; >> margin-top: 15px; >> margin-bottom: 10px; >> border: 1px solid #99CCFF; >> } >> >> Background: >> If no width definition for the Box exist, so Firefox use the minimum needed width, IE6 the maximum available width. >> > > Okay, I've reverted r19832 in r20191 :) > Sorry for that, I had tested in IE but just IE 7 because you can't install two of them (at least not without some registry hacking). >> Problems: >> If the box has a fixed width definition, it will not resize with the font zooming feature. >> > > Well, I guess we will have to cope with that for the time being :/ > I'm still working on the drupal thingie, but I'm a little short on time currently. One of the drupal coders, drewish, was so kind to provide a plugin which fixes the tranlation problems http://drupal.org/project/active_translation -- Follow the blue rabbit - The Freenet Project - http://freenetproject.org/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 252 bytes Desc: OpenPGP digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/a428e4be/attachment.pgp From Lokivgap at gmx.de Wed Jun 4 09:50:59 2008 From: Lokivgap at gmx.de (Loki) Date: Wed, 4 Jun 2008 11:50:59 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <20080604084028.GI3506@freenetproject.org> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> <20080604084028.GI3506@freenetproject.org> Message-ID: <20080604115059.1ae1de43.Lokivgap@gmx.de> On Wed, 4 Jun 2008 10:40:30 +0200 Florent Daigni?re wrote: > > Background: > > If no width definition for the Box exist, so Firefox use the minimum needed width, IE6 the maximum available width. > > > > Okay, I've reverted r19832 in r20191 :) > > > Problems: > > If the box has a fixed width definition, it will not resize with the font zooming feature. > > > > Well, I guess we will have to cope with that for the time being :/ > I read somewhere the the IE7 at last support "min-width" definition. I have only a IE6 on WIN2000 test installation running, so i can't test this. This would be the solution for the font zooming problem, but not for the IE6 users. From toad at amphibian.dyndns.org Wed Jun 4 11:32:17 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Wed, 4 Jun 2008 12:32:17 +0100 Subject: [freenet-dev] Website and IE In-Reply-To: <20080604084028.GI3506@freenetproject.org> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> <20080604084028.GI3506@freenetproject.org> Message-ID: <200806041232.17845.toad@amphibian.dyndns.org> On Wednesday 04 June 2008 09:40, Florent Daigni?re wrote: > * Loki [2008-06-04 10:13:51]: > > > Well not a wizard here, but maybe a help. > > > > In style.css just activate the out commented line /*width: 150px;*/ > > > > div#menu { > > font-family: Sans-serif, Arial, "Times New Roman", Georgia, Times, serif; > > float:left; > > /*width: 150px;*/ <---- > > font-style: normal; > > font-weight: normal; > > text-align: left; > > line-height: 100%; > > color: #000000; > > background: #CCDDFF; > > text-decoration: none; > > margin-top: 15px; > > margin-bottom: 10px; > > border: 1px solid #99CCFF; > > } > > > > Background: > > If no width definition for the Box exist, so Firefox use the minimum needed width, IE6 the maximum available width. > > > > Okay, I've reverted r19832 in r20191 :) > > > Problems: > > If the box has a fixed width definition, it will not resize with the font zooming feature. > > > > Well, I guess we will have to cope with that for the time being :/ > Can't we make it use a font-relative size? IIRC CSS supports something like that? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/918d5a59/attachment.pgp From toad at amphibian.dyndns.org Wed Jun 4 11:33:40 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Wed, 4 Jun 2008 12:33:40 +0100 Subject: [freenet-dev] =?utf-8?q?Fwd=3A_=5Bfreenet-support=5D_Usability=09?= =?utf-8?q?testing=3Fon=3FWindows_XP?= In-Reply-To: <20080604043550.GD3506@freenetproject.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806031722.56480.toad@amphibian.dyndns.org> <20080604043550.GD3506@freenetproject.org> Message-ID: <200806041233.41459.toad@amphibian.dyndns.org> On Wednesday 04 June 2008 05:35, Florent Daigni?re wrote: > * Matthew Toseland [2008-06-03 17:22:56]: > > > On Monday 02 June 2008 17:37, Julien Cornuwel wrote: > > > Florent Daigni?re a ?crit : > > > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > > > >>> Additionnaly, that page was in English despite the fact he chose French > > > >>> in the installer. I had to tell him he had to finish the installer and > > > >>> then, click on the link to the first-time wizard. > > > >> Okay this is bad, I think it's been fixed though. Nextgens? > > > >> > > > > > > > > Yeah, it was a windows-specific bug and which has been fixed. > > > > > > Before or after I did the test ? Because it was not at that moment. > > > > > > >>> - Just a detail : every shortcuts (except Browse Freenet) in the menu > > > >>> came with the same default icon. He was disapointed and I had to tell > > > >>> him it was irrelevant and everything was fine. > > > >>> > > > > > > > > If you have icon files I can use I'll be pleased to fix that > > > > > > Well, no I don't *have* any. And seriously, you don't want me to draw > > > some ;) > > > > > > >>> Maybe we should only spawn the browser at the very end of the > > > >>> installation process, when the installer closes and Freenet is up and > > > >>> running. > > > >> We open the browser very shortly after Freenet is working. However, at > > the > > > >> moment, we download and install more stuff after that - namely the > > bundled > > > >> apps. In any case, without major changes to the installer, we have to > > open > > > >> the browser during the processes page (again, am I right nextgens? Is > > there > > > >> any possibility of exit-hooks?). Is it a big deal? > > > > > > > > Last time I tried to use them they were issues > > > > > > > >> The most likely consequence is that the user doesn't complete the > > installer, > > > >> or that there is a long delay before it, right? > > > >> > > > > > > > > We could time-delay the browser startup if that's the only thing :) > > > > > > Any solution would be good. Small details like that keep users telling > > > others that Freenet is complicated. We are so used to it that we can't > > > see them anymore but, newbies do... > > > > I'm not sure that a time delay would help. And exit hooks don't work. So what > > can we do? > > Maybe we should not open the browser and wait for the user to click on > "browse-freenet". > I dunno... what exactly is it that confuses some users? Most installers for e.g. games have a checkbox on the last page for whether to open the app you just installed on exiting the installer, really we should do that. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/22974e0f/attachment.pgp From toad at amphibian.dyndns.org Wed Jun 4 11:34:58 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Wed, 4 Jun 2008 12:34:58 +0100 Subject: [freenet-dev] =?utf-8?q?Fwd=3A_=5Bfreenet-support=5D_Usability=09?= =?utf-8?q?testing=3Fon=3FWindows_XP?= In-Reply-To: <20080604044216.GE3506@freenetproject.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806031724.48961.toad@amphibian.dyndns.org> <20080604044216.GE3506@freenetproject.org> Message-ID: <200806041235.00654.toad@amphibian.dyndns.org> On Wednesday 04 June 2008 05:42, Florent Daigni?re wrote: > * Matthew Toseland [2008-06-03 17:24:48]: > > > On Monday 02 June 2008 15:13, Florent Daigni?re wrote: > > > * Matthew Toseland [2008-06-02 15:03:37]: > > > > > > > On Monday 02 June 2008 14:47, Florent Daigni?re wrote: > > > > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > > > > > > > Okay, what issues do we have here? > > > > > > > > > > > > 1. The processes page takes longer than the copying files page. In > > fact, > > > > most > > > > > > of the time of the installer is occupied by the processes page. > > > > > > > > > > > > IMHO the solution to this is to bundle everything which is essential > > with > > > > the > > > > > > base installer. Then the processes page will take much less time. Also > > we > > > > > > won't need a separate bundle-installer, so it will be easier to use in > > > > places > > > > > > where the Freenet website is blocked. The downside is the installer > > will > > > > be > > > > > > significantly bigger. > > > > > > > > > > I'm not convinced about that > > > > > > > > Why? It won't be huge, if we don't bundle any unnecessary Huge Stuff (see > > > > below). And it's not small now. > > > > > > That means we will have to generate new installers on a regular basis... > > > I don't like that :p > > > > Is there any good reason for this dislike? > > Downloading stuffs during the install process actually tests the user's > connectivity; If we don't do that we will end up with more support > requests from users having non-freenet-related issues. > > Moreover, if we want the "trust-chain" in between us and the newbies to > be established, we HAVE TO distribute the installer from emu and not > from mirrors... meaning that it has to be as small as possible if we > intend to stand a slashdoting. > > Arguably we don't distribute the installer ourselves yet... but I was in > the process of changing that. > Well then we'll have to make it a lot smaller than even 3.6MB: we'd want some sort of minimal shell that downloads a hash from emu and then fetches the rest from a mirror. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/da709e0b/attachment.pgp From nextgens at freenetproject.org Wed Jun 4 11:35:19 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 13:35:19 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <48465EB4.9040005@users.sourceforge.net> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> <20080604084028.GI3506@freenetproject.org> <48465EB4.9040005@users.sourceforge.net> Message-ID: <20080604113517.GK3506@freenetproject.org> * Michael T?nzer [2008-06-04 11:21:56]: > I'm still working on the drupal thingie, but I'm a little short on time > currently. > One of the drupal coders, drewish, was so kind to provide a plugin which > fixes the tranlation problems http://drupal.org/project/active_translation I have set it up on the vhost... but it seems to be encountering problems rebuilding the translation table -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/389b5d55/attachment.pgp From nextgens at freenetproject.org Wed Jun 4 11:36:21 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 13:36:21 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability?testing?on?Windows XP In-Reply-To: <200806041233.41459.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806031722.56480.toad@amphibian.dyndns.org> <20080604043550.GD3506@freenetproject.org> <200806041233.41459.toad@amphibian.dyndns.org> Message-ID: <20080604113619.GL3506@freenetproject.org> * Matthew Toseland [2008-06-04 12:33:40]: > On Wednesday 04 June 2008 05:35, Florent Daigni?re wrote: > > * Matthew Toseland [2008-06-03 17:22:56]: > > > > > On Monday 02 June 2008 17:37, Julien Cornuwel wrote: > > > > Florent Daigni?re a ?crit : > > > > > * Matthew Toseland [2008-06-02 14:15:02]: > > > > > > > > > >>> Additionnaly, that page was in English despite the fact he chose > French > > > > >>> in the installer. I had to tell him he had to finish the installer > and > > > > >>> then, click on the link to the first-time wizard. > > > > >> Okay this is bad, I think it's been fixed though. Nextgens? > > > > >> > > > > > > > > > > Yeah, it was a windows-specific bug and which has been fixed. > > > > > > > > Before or after I did the test ? Because it was not at that moment. > > > > > > > > >>> - Just a detail : every shortcuts (except Browse Freenet) in the > menu > > > > >>> came with the same default icon. He was disapointed and I had to > tell > > > > >>> him it was irrelevant and everything was fine. > > > > >>> > > > > > > > > > > If you have icon files I can use I'll be pleased to fix that > > > > > > > > Well, no I don't *have* any. And seriously, you don't want me to draw > > > > some ;) > > > > > > > > >>> Maybe we should only spawn the browser at the very end of the > > > > >>> installation process, when the installer closes and Freenet is up > and > > > > >>> running. > > > > >> We open the browser very shortly after Freenet is working. However, > at > > > the > > > > >> moment, we download and install more stuff after that - namely the > > > bundled > > > > >> apps. In any case, without major changes to the installer, we have to > > > open > > > > >> the browser during the processes page (again, am I right nextgens? Is > > > there > > > > >> any possibility of exit-hooks?). Is it a big deal? > > > > > > > > > > Last time I tried to use them they were issues > > > > > > > > > >> The most likely consequence is that the user doesn't complete the > > > installer, > > > > >> or that there is a long delay before it, right? > > > > >> > > > > > > > > > > We could time-delay the browser startup if that's the only thing :) > > > > > > > > Any solution would be good. Small details like that keep users telling > > > > others that Freenet is complicated. We are so used to it that we can't > > > > see them anymore but, newbies do... > > > > > > I'm not sure that a time delay would help. And exit hooks don't work. So > what > > > can we do? > > > > Maybe we should not open the browser and wait for the user to click on > > "browse-freenet". > > > I dunno... what exactly is it that confuses some users? > > Most installers for e.g. games have a checkbox on the last page for whether to > open the app you just installed on exiting the installer, really we should do > that. That's not trivial -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/a13f4235/attachment.pgp From twh25 at cam.ac.uk Wed Jun 4 12:18:11 2008 From: twh25 at cam.ac.uk (Theodore Hong) Date: Wed, 4 Jun 2008 13:18:11 +0100 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing?on?Windows XP In-Reply-To: <200806041235.00654.toad@amphibian.dyndns.org> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806031724.48961.toad@amphibian.dyndns.org> <20080604044216.GE3506@freenetproject.org> <200806041235.00654.toad@amphibian.dyndns.org> Message-ID: <179050340806040518w10377fd7y6e7b479318e60b01@mail.gmail.com> Matthew Toseland wrote: > On Wednesday 04 June 2008 05:42, Florent Daigni?re wrote: >> Moreover, if we want the "trust-chain" in between us and the newbies to >> be established, we HAVE TO distribute the installer from emu and not >> from mirrors... meaning that it has to be as small as possible if we >> intend to stand a slashdoting. >> > Well then we'll have to make it a lot smaller than even 3.6MB: we'd want some > sort of minimal shell that downloads a hash from emu and then fetches the > rest from a mirror. Is there a project signing key that we can use to sign releases? theo From nextgens at freenetproject.org Wed Jun 4 12:20:29 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 14:20:29 +0200 Subject: [freenet-dev] Fwd: [freenet-support] Usability testing?on?Windows XP In-Reply-To: <179050340806040518w10377fd7y6e7b479318e60b01@mail.gmail.com> References: <200805301757.49729.toad@amphibian.dyndns.org> <200806031724.48961.toad@amphibian.dyndns.org> <20080604044216.GE3506@freenetproject.org> <200806041235.00654.toad@amphibian.dyndns.org> <179050340806040518w10377fd7y6e7b479318e60b01@mail.gmail.com> Message-ID: <20080604122027.GM3506@freenetproject.org> * Theodore Hong [2008-06-04 13:18:11]: > Matthew Toseland wrote: > > On Wednesday 04 June 2008 05:42, Florent Daigni?re wrote: > >> Moreover, if we want the "trust-chain" in between us and the newbies to > >> be established, we HAVE TO distribute the installer from emu and not > >> from mirrors... meaning that it has to be as small as possible if we > >> intend to stand a slashdoting. > >> > > Well then we'll have to make it a lot smaller than even 3.6MB: we'd want some > > sort of minimal shell that downloads a hash from emu and then fetches the > > rest from a mirror. > > Is there a project signing key that we can use to sign releases? > theo http://archives.freenetproject.org/message/20080515.125656.0d8a973e.en.html For X509 (jarsigner) and http://downloads.freenetproject.org/alpha/README for GPG -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/3846c2d4/attachment.pgp From Lokivgap at gmx.de Wed Jun 4 13:02:27 2008 From: Lokivgap at gmx.de (Loki) Date: Wed, 4 Jun 2008 15:02:27 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <200806041232.17845.toad@amphibian.dyndns.org> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> <20080604084028.GI3506@freenetproject.org> <200806041232.17845.toad@amphibian.dyndns.org> Message-ID: <20080604150227.f22d0967.Lokivgap@gmx.de> On Wed, 4 Jun 2008 12:32:17 +0100 Matthew Toseland wrote: > > > > > > > Well, I guess we will have to cope with that for the time being :/ > > > Can't we make it use a font-relative size? IIRC CSS supports something like > that? You are right! "width: 9.5em;" should do the magic! From nextgens at freenetproject.org Wed Jun 4 13:30:57 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Wed, 4 Jun 2008 15:30:57 +0200 Subject: [freenet-dev] Website and IE In-Reply-To: <20080604150227.f22d0967.Lokivgap@gmx.de> References: <20080604045810.GG3506@freenetproject.org> <20080604101351.ea19e51a.Lokivgap@gmx.de> <20080604084028.GI3506@freenetproject.org> <200806041232.17845.toad@amphibian.dyndns.org> <20080604150227.f22d0967.Lokivgap@gmx.de> Message-ID: <20080604133055.GN3506@freenetproject.org> * Loki [2008-06-04 15:02:27]: > On Wed, 4 Jun 2008 12:32:17 +0100 Matthew Toseland wrote: > > > > > > > > > > > Well, I guess we will have to cope with that for the time being :/ > > > > > Can't we make it use a font-relative size? IIRC CSS supports something like > > that? > > You are right! "width: 9.5em;" should do the magic! > Commited in r20198; It's not perfect but it's good enough :) -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080604/d5e63278/attachment.pgp From bombe at pterodactylus.net Thu Jun 5 08:49:57 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 10:49:57 +0200 Subject: [freenet-dev] git-over-fproxy Message-ID: <200806051049.58509.bombe@pterodactylus.net> Hey, everyone. I tried to get git over fproxy working last night and I got a couple of results. I was able to insert a cloned repository into freenet and retrieve it with the following URL: http://localhost:8888/USK at .../foo.git/2?type=text/plain The reason for this is the warning page that fproxy shows when it does not know the content type of a file or does not know how to filter it. Adding a command-line parameter right now is the only way to make fproxy deliver the content of the file immediately. Then fproxy also insisted on returning HTTP status code 500 if a file is not in a site?s archive but that was changed easily (r20207). After I fixed up git to include the path of the objects it wants into the URL before the query string I was able to successfully pull a repository from freenet. I though I could just roll up the patch and submit it to git but there were indeed some problems with it (as pointed out[1] by Junio). Basically, my patch will break URLs like http://foo.org/serve.cgi?project=bar.git. That is pretty much unacceptable. So I need a way to retrieve a file from freenet _without_ the warning page and _without_ adding a query string to the URL. Can we introduce something like http://localhost:8888/unchecked/ to retrieve the file, no matter what? Maybe with a config option to disable that behaviour to prevent people from clicking the link and being ?infected? with whatever is delivered? As I'm starting to like git more and more I do want to see it supported in Freenet and this is the only way to make it work that I can currently think about. Thank you for listening, David [1] http://article.gmane.org/gmane.comp.version-control.git/83875 -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/c872c9d9/attachment.pgp From j16sdiz+freenet at gmail.com Thu Jun 5 09:04:49 2008 From: j16sdiz+freenet at gmail.com (Daniel Cheng) Date: Thu, 5 Jun 2008 17:04:49 +0800 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051049.58509.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> Message-ID: On Thu, Jun 5, 2008 at 4:49 PM, David 'Bombe' Roden wrote: > Hey, everyone. > > I tried to get git over fproxy working last night and I got a couple of > results. As I have point out in FMS, If you pack all files in the same archive: this disallow incremental updates to the repository. If you don't: git "walks" the DAG of blobs. would spend lots of time. > I was able to insert a cloned repository into freenet and retrieve it with the > following URL: > > http://localhost:8888/USK at .../foo.git/2?type=text/plain > > The reason for this is the warning page that fproxy shows when it does not > know the content type of a file or does not know how to filter it. Adding a > command-line parameter right now is the only way to make fproxy deliver the > content of the file immediately. > > Then fproxy also insisted on returning HTTP status code 500 if a file is not > in a site's archive but that was changed easily (r20207). > > After I fixed up git to include the path of the objects it wants into the URL > before the query string I was able to successfully pull a repository from > freenet. > > I though I could just roll up the patch and submit it to git but there were > indeed some problems with it (as pointed out[1] by Junio). Basically, my > patch will break URLs like http://foo.org/serve.cgi?project=bar.git. That is > pretty much unacceptable. > > So I need a way to retrieve a file from freenet _without_ the warning page and > _without_ adding a query string to the URL. Can we introduce something like > http://localhost:8888/unchecked/ to retrieve the file, no matter what? > Maybe with a config option to disable that behaviour to prevent people from > clicking the link and being "infected" with whatever is delivered? Alternatively, making fproxy a real HTTP proxy server. All traffics, include those sent to internet, are pass thru the proxy server. In this way, we don't have to worry about external url in and friends and hence mitigating the need to sanitize the HTML. > As I'm starting to like git more and more I do want to see it supported in > Freenet and this is the only way to make it work that I can currently think > about. > > > Thank you for listening, > > David > > [1] http://article.gmane.org/gmane.comp.version-control.git/83875 > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > From bombe at pterodactylus.net Thu Jun 5 09:33:18 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 11:33:18 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> Message-ID: <200806051133.22304.bombe@pterodactylus.net> On Thursday 05 June 2008 11:04:49 Daniel Cheng wrote: > git "walks" the DAG of blobs. would spend lots of time. Actually it's not much slower than "normal" http because all the object blobs are within the same container of the site, for small repositories at least. Also, you can git-prune and git-gc your repository before uploading. That will create a pack containing all objects, which will be a lot faster than walking all objects. > If you pack all files in the same archive: > this disallow incremental updates to the repository. We also don't have incremental updates for freesites, so what? > Alternatively, making fproxy a real HTTP proxy server. > All traffics, include those sent to internet, are pass thru the proxy > server. fproxy, though it has a "proxy" in its name, is not a proxy. It's an http-gateway to freenet. I, personally, would like to keep it that way. :) David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/c66bb1c3/attachment.pgp From bombe at pterodactylus.net Thu Jun 5 09:43:07 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 11:43:07 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> Message-ID: <200806051143.08258.bombe@pterodactylus.net> On Thursday 05 June 2008 11:04:49 Daniel Cheng wrote: > Alternatively, making fproxy a real HTTP proxy server. As I have just been told by nextgens fproxy already _is_ an http proxy. I confirmed just two minutes ago, it can be used as a proxy which will only deliver freenet stuff. But it still has the warning page for unknown content types which makes it unusable for what I want. :/ David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/70f8db34/attachment.pgp From bbackde at googlemail.com Thu Jun 5 10:17:05 2008 From: bbackde at googlemail.com (bbackde at googlemail.com) Date: Thu, 5 Jun 2008 12:17:05 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051049.58509.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> Message-ID: Why git, why not mercurial (better multi-platform support than git, same or more features)? (maybe you read the article about mercurial in the latest c't?) On Thu, Jun 5, 2008 at 10:49 AM, David 'Bombe' Roden wrote: > Hey, everyone. > > I tried to get git over fproxy working last night and I got a couple of > results. > > I was able to insert a cloned repository into freenet and retrieve it with the > following URL: > > http://localhost:8888/USK at .../foo.git/2?type=text/plain > > The reason for this is the warning page that fproxy shows when it does not > know the content type of a file or does not know how to filter it. Adding a > command-line parameter right now is the only way to make fproxy deliver the > content of the file immediately. > > Then fproxy also insisted on returning HTTP status code 500 if a file is not > in a site's archive but that was changed easily (r20207). > > After I fixed up git to include the path of the objects it wants into the URL > before the query string I was able to successfully pull a repository from > freenet. > > I though I could just roll up the patch and submit it to git but there were > indeed some problems with it (as pointed out[1] by Junio). Basically, my > patch will break URLs like http://foo.org/serve.cgi?project=bar.git. That is > pretty much unacceptable. > > So I need a way to retrieve a file from freenet _without_ the warning page and > _without_ adding a query string to the URL. Can we introduce something like > http://localhost:8888/unchecked/ to retrieve the file, no matter what? > Maybe with a config option to disable that behaviour to prevent people from > clicking the link and being "infected" with whatever is delivered? > > As I'm starting to like git more and more I do want to see it supported in > Freenet and this is the only way to make it work that I can currently think > about. > > > Thank you for listening, > > David > > [1] http://article.gmane.org/gmane.comp.version-control.git/83875 > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > -- __________________________________________________ GnuPG key: (0x48DBFA8A) Keyserver: pgpkeys.pca.dfn.de Fingerprint: 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A __________________________________________________ From bombe at pterodactylus.net Thu Jun 5 10:28:57 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 12:28:57 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051049.58509.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> Message-ID: <200806051229.00191.bombe@pterodactylus.net> On Thursday 05 June 2008 10:49:57 David ?Bombe? Roden wrote: > So I need a way to retrieve a file from freenet _without_ the warning page > and _without_ adding a query string to the URL. Can we introduce something > like http://localhost:8888/unchecked/ to retrieve the file, no matter > what? Turns out it?s already almost in fproxy. By using "http://node:8888/?key=" a redirect to "http://node:8888/" is issued, and as curl follows redirects transparently this would work, if a given content type would also be used in the redirect target URL. Index: src/freenet/clients/http/FProxyToadlet.java =================================================================== --- src/freenet/clients/http/FProxyToadlet.java (revision 20211) +++ src/freenet/clients/http/FProxyToadlet.java (working copy) @@ -336,7 +336,12 @@ } if(logMINOR) Logger.minor(this, "Redirecting to FreenetURI: "+newURI); - headers.put("Location", "/"+newURI); + String type = httprequest.getParam("type"); + if (type != null) { + headers.put("Location", "/" + newURI + "?type=" + type); + } else { + headers.put("Location", "/" + newURI); + } ctx.sendReplyHeaders(302, "Found", headers, null, 0); return; } Guess what this does. :) IMHO this patch is completely harmless: it allows nothing fproxy didn't already do as also right now you can simply append "type=text/plain" to your URL and thus skip the content filter. This patch would also allow git to work with freenet without any modifications what-so-ever. David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/381db015/attachment.pgp From bombe at pterodactylus.net Thu Jun 5 10:34:13 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 12:34:13 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> Message-ID: <200806051234.13574.bombe@pterodactylus.net> On Thursday 05 June 2008 12:17:05 bbackde at googlemail.com wrote: > Why git, why not mercurial (better multi-platform support than git, > same or more features)? Why mercurial, why not git? This kind of question is pointless. And annoying. Especially annoying. David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/a01bac3f/attachment.pgp From j16sdiz+freenet at gmail.com Thu Jun 5 10:58:56 2008 From: j16sdiz+freenet at gmail.com (Daniel Cheng) Date: Thu, 5 Jun 2008 18:58:56 +0800 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051133.22304.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> <200806051133.22304.bombe@pterodactylus.net> Message-ID: On Thu, Jun 5, 2008 at 5:33 PM, David 'Bombe' Roden wrote: > On Thursday 05 June 2008 11:04:49 Daniel Cheng wrote: > >> git "walks" the DAG of blobs. would spend lots of time. > > Actually it's not much slower than "normal" http because all the object blobs > are within the same container of the site, for small repositories at least. > > Also, you can git-prune and git-gc your repository before uploading. That will > create a pack containing all objects, which will be a lot faster than walking > all objects. I am aware of this, ... but... (see below.) >> If you pack all files in the same archive: >> this disallow incremental updates to the repository. > > We also don't have incremental updates for freesites, so what? Unlike most websites, source repository size grow with time. Also, git does not have fully-functional partial clone. This means the upload / download size would eventally bite us. >> Alternatively, making fproxy a real HTTP proxy server. >> All traffics, include those sent to internet, are pass thru the proxy >> server. > > fproxy, though it has a "proxy" in its name, is not a proxy. It's an > http-gateway to freenet. I, personally, would like to keep it that way. :) > > > David > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > From toad at amphibian.dyndns.org Thu Jun 5 11:55:44 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Thu, 5 Jun 2008 12:55:44 +0100 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051234.13574.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> <200806051234.13574.bombe@pterodactylus.net> Message-ID: <200806051255.49367.toad@amphibian.dyndns.org> On Thursday 05 June 2008 11:34, David ?Bombe? Roden wrote: > On Thursday 05 June 2008 12:17:05 bbackde at googlemail.com wrote: > > > Why git, why not mercurial (better multi-platform support than git, > > same or more features)? > > Why mercurial, why not git? > > This kind of question is pointless. And annoying. Especially annoying. We should have all of them! But we need to have at least one working well. Right now even the mercurial support is incomplete. > > > David > -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/7de3e149/attachment.pgp From edt at aei.ca Thu Jun 5 15:00:11 2008 From: edt at aei.ca (Ed Tomlinson) Date: Thu, 5 Jun 2008 11:00:11 -0400 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051234.13574.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> <200806051234.13574.bombe@pterodactylus.net> Message-ID: <200806051100.12187.edt@aei.ca> On June 5, 2008, David ?Bombe? Roden wrote: > On Thursday 05 June 2008 12:17:05 bbackde at googlemail.com wrote: > > > Why git, why not mercurial (better multi-platform support than git, > > same or more features)? > > Why mercurial, why not git? > > This kind of question is pointless. And annoying. Especially annoying. David, This is the best reply I have seen to this sort of 'supid' message in along time. Ed From bbackde at googlemail.com Thu Jun 5 15:41:05 2008 From: bbackde at googlemail.com (bbackde at googlemail.com) Date: Thu, 5 Jun 2008 17:41:05 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051100.12187.edt@aei.ca> References: <200806051049.58509.bombe@pterodactylus.net> <200806051234.13574.bombe@pterodactylus.net> <200806051100.12187.edt@aei.ca> Message-ID: One final word to this: I don't wanted to start a flamewar, I wanted to point out that one could concentrate on a system that works well on all platforms. This is not true for git. But ok, go on and do it at least for the Linux users... Btw: I gave two reasons for my recommendation of mercurial, but David just ignored them. For me his reply was not the best, but annoying. On Thu, Jun 5, 2008 at 5:00 PM, Ed Tomlinson wrote: > On June 5, 2008, David 'Bombe' Roden wrote: >> On Thursday 05 June 2008 12:17:05 bbackde at googlemail.com wrote: >> >> > Why git, why not mercurial (better multi-platform support than git, >> > same or more features)? >> >> Why mercurial, why not git? >> >> This kind of question is pointless. And annoying. Especially annoying. > > David, > > This is the best reply I have seen to this sort of 'supid' message in along time. > > > Ed > > > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl -- __________________________________________________ GnuPG key: (0x48DBFA8A) Keyserver: pgpkeys.pca.dfn.de Fingerprint: 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A __________________________________________________ From j16sdiz+freenet at gmail.com Thu Jun 5 16:04:10 2008 From: j16sdiz+freenet at gmail.com (Daniel Cheng) Date: Fri, 6 Jun 2008 00:04:10 +0800 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> <200806051234.13574.bombe@pterodactylus.net> <200806051100.12187.edt@aei.ca> Message-ID: On Thu, Jun 5, 2008 at 11:41 PM, wrote: > One final word to this: I don't wanted to start a flamewar, I wanted > to point out that > one could concentrate on a system that works well on all platforms. > > This is not true for git. But ok, go on and do it at least for the > Linux users... Not true. There is a java port of git, which is faster then early git core. The native windows port works, although no quite as fast. Mercurial depends on the python chains. for ad-hoc scripts Git is designed for quick one-liner using shell scripts. Mercurial require python knowledge to do the simplest automation. > Btw: I gave two reasons for my recommendation of mercurial, but David > just ignored them. > For me his reply was not the best, but annoying. > > On Thu, Jun 5, 2008 at 5:00 PM, Ed Tomlinson wrote: >> On June 5, 2008, David 'Bombe' Roden wrote: >>> On Thursday 05 June 2008 12:17:05 bbackde at googlemail.com wrote: >>> >>> > Why git, why not mercurial (better multi-platform support than git, >>> > same or more features)? >>> >>> Why mercurial, why not git? >>> >>> This kind of question is pointless. And annoying. Especially annoying. >> >> David, >> >> This is the best reply I have seen to this sort of 'supid' message in along time. >> >> >> Ed >> >> >> >> _______________________________________________ >> Devl mailing list >> Devl at freenetproject.org >> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > > > > -- > __________________________________________________ > GnuPG key: (0x48DBFA8A) > Keyserver: pgpkeys.pca.dfn.de > Fingerprint: > 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A > __________________________________________________ > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > From bbackde at googlemail.com Thu Jun 5 16:12:30 2008 From: bbackde at googlemail.com (bbackde at googlemail.com) Date: Thu, 5 Jun 2008 18:12:30 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> <200806051234.13574.bombe@pterodactylus.net> <200806051100.12187.edt@aei.ca> Message-ID: Ok. Finally there is some constructive discussion. Great. On Thu, Jun 5, 2008 at 6:04 PM, Daniel Cheng wrote: > On Thu, Jun 5, 2008 at 11:41 PM, wrote: >> One final word to this: I don't wanted to start a flamewar, I wanted >> to point out that >> one could concentrate on a system that works well on all platforms. >> >> This is not true for git. But ok, go on and do it at least for the >> Linux users... > > Not true. There is a java port of git, which is faster then early git core. > The native windows port works, although no quite as fast. > > Mercurial depends on the python chains. > > for ad-hoc scripts > Git is designed for quick one-liner using shell scripts. > Mercurial require python knowledge to do the simplest automation. > >> Btw: I gave two reasons for my recommendation of mercurial, but David >> just ignored them. >> For me his reply was not the best, but annoying. >> >> On Thu, Jun 5, 2008 at 5:00 PM, Ed Tomlinson wrote: >>> On June 5, 2008, David 'Bombe' Roden wrote: >>>> On Thursday 05 June 2008 12:17:05 bbackde at googlemail.com wrote: >>>> >>>> > Why git, why not mercurial (better multi-platform support than git, >>>> > same or more features)? >>>> >>>> Why mercurial, why not git? >>>> >>>> This kind of question is pointless. And annoying. Especially annoying. >>> >>> David, >>> >>> This is the best reply I have seen to this sort of 'supid' message in along time. >>> >>> >>> Ed >>> >>> >>> >>> _______________________________________________ >>> Devl mailing list >>> Devl at freenetproject.org >>> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl >> >> >> >> -- >> __________________________________________________ >> GnuPG key: (0x48DBFA8A) >> Keyserver: pgpkeys.pca.dfn.de >> Fingerprint: >> 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A >> __________________________________________________ >> _______________________________________________ >> Devl mailing list >> Devl at freenetproject.org >> http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl >> > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > -- __________________________________________________ GnuPG key: (0x48DBFA8A) Keyserver: pgpkeys.pca.dfn.de Fingerprint: 477D F057 1BD4 1AE7 8A54 8679 6690 E2EC 48DB FA8A __________________________________________________ From bombe at pterodactylus.net Thu Jun 5 16:38:10 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 18:38:10 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> <200806051100.12187.edt@aei.ca> Message-ID: <200806051838.12600.bombe@pterodactylus.net> On Thursday 05 June 2008 17:41:05 bbackde at googlemail.com wrote: > Btw: I gave two reasons for my recommendation of mercurial, but David > just ignored them. Yes, I did. And why? Because a) the SCM I use for my projects is my choice. And I don?t make my choice by looking at which platforms are supported by the SCM. (I couldn?t care less about e.g. Windows. :) b) I am already used to git. Why should switch to another SCM if git meets all my requirements? Heck, even subversion meets (almost) all my requirements because those are quite modest, and the only thing missing from subversion is a possibility to have the repository in freenet. c) that kind of question _is_ annoying. And pointless. Damn, d?j?-vu. No, the point is that this kind of question creates flamewars easier than anything else which is why I didn?t bother to consider your arguments. d) I don?t think that one working tool for a specific task is enough. For me all this open source stuff is about choice, and I want to have more choice than ?use it or don?t?. David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/bd508db8/attachment.pgp From bbackde at googlemail.com Thu Jun 5 16:59:04 2008 From: bbackde at googlemail.com (bbackde at googlemail.com) Date: Thu, 5 Jun 2008 18:59:04 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806051838.12600.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> <200806051100.12187.edt@aei.ca> <200806051838.12600.bombe@pterodactylus.net> Message-ID: On Thu, Jun 5, 2008 at 6:38 PM, David 'Bombe' Roden wrote: > On Thursday 05 June 2008 17:41:05 bbackde at googlemail.com wrote: > >> Btw: I gave two reasons for my recommendation of mercurial, but David >> just ignored them. > > Yes, I did. And why? Because > > a) the SCM I use for my projects is my choice. And I don't make my choice by > looking at which platforms are supported by the SCM. (I couldn't care less > about e.g. Windows. :) Full ack. > > b) I am already used to git. Why should switch to another SCM if git meets all > my requirements? Heck, even subversion meets (almost) all my requirements > because those are quite modest, and the only thing missing from subversion is > a possibility to have the repository in freenet. Ok for me. I also use and like svn so far. > > c) that kind of question _is_ annoying. And pointless. Damn, d?j?-vu. No, the > point is that this kind of question creates flamewars easier than anything > else which is why I didn't bother to consider your arguments. My point (and I was misunderstood) was: if someone works on such a 'port' for freenet, it would be great if it would fit for all users, not only for the few who use Linux (*g*). But now I read that git fits well also for Windows users, so all is fine for me. > > d) I don't think that one working tool for a specific task is enough. For me > all this open source stuff is about choice, and I want to have more choice > than "use it or don't". Thats true. And it would be great if there would be at least one fully working solution, not a few half-working ones .... you have the chance to create one. Thanks for that. > > > David > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > From bombe at pterodactylus.net Thu Jun 5 18:03:16 2008 From: bombe at pterodactylus.net (David =?utf-8?q?=E2=80=98Bombe=E2=80=99_Roden?=) Date: Thu, 5 Jun 2008 20:03:16 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: References: <200806051049.58509.bombe@pterodactylus.net> <200806051838.12600.bombe@pterodactylus.net> Message-ID: <200806052003.20462.bombe@pterodactylus.net> On Thursday 05 June 2008 18:59:04 bbackde at googlemail.com wrote: > My point (and I was misunderstood) was: if someone works on such a > 'port' for freenet, it would be great if it would fit for all users, not > only for the few who use Linux (*g*). The platforms that are supported by Git are not really something I can do about? > But now I read that git fits well also for Windows users, so all is fine > for me. ?but others seem to have taken care of that. Wonderful! :) > Thats true. And it would be great if there would be at least one fully > working solution, not a few half-working ones .... you have the chance to > create one. Daniel mentioned a thing on IRC that needs to be taken care of when the time arrives: repositories of course grow over time. A repository might get big enough that inserting it will become a pain in the ass; unfortunately, the only kind of ?incremental update? that we could think of does have other drawbacks, e.g. disappearing data. I?m not completely sure what to do about that. David -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 197 bytes Desc: This is a digitally signed message part. Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/a3b919f6/attachment.pgp From NEOatNHNG at users.sourceforge.net Thu Jun 5 20:41:13 2008 From: NEOatNHNG at users.sourceforge.net (=?UTF-8?B?TWljaGFlbCBUw6RuemVy?=) Date: Thu, 05 Jun 2008 22:41:13 +0200 Subject: [freenet-dev] git-over-fproxy In-Reply-To: <200806052003.20462.bombe@pterodactylus.net> References: <200806051049.58509.bombe@pterodactylus.net> <200806051838.12600.bombe@pterodactylus.net> <200806052003.20462.bombe@pterodactylus.net> Message-ID: <48484F69.4040801@users.sourceforge.net> David ?Bombe? Roden schrieb: > > Daniel mentioned a thing on IRC that needs to be taken care of when the time > arrives: repositories of course grow over time. A repository might get big > enough that inserting it will become a pain in the ass; unfortunately, the > only kind of ?incremental update? that we could think of does have other > drawbacks, e.g. disappearing data. I?m not completely sure what to do about > that. > How about inserting it incremental but every Xth (or if delta(t)>x) commit a full version of the repository is released? So every checkout needs the last full version, plus the incremental commits, which have been released since then, an update (if it's somewhat up to date) only needs to pull the incremental commits. regards Neo at NHNG -- Follow the blue rabbit - The Freenet Project - http://freenetproject.org/ -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 252 bytes Desc: OpenPGP digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080605/f3dbf66a/attachment.pgp From nextgens at freenetproject.org Fri Jun 6 03:18:43 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Fri, 6 Jun 2008 05:18:43 +0200 Subject: [freenet-dev] [freenet-cvs] r20222 - in trunk/apps/new_installer: . res/firefox_profile res/firefox_profile/extensions res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a } res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a }/chrome In-Reply-To: <20080605164830.9674D4797DB@freenetproject.org> References: <20080605164830.9674D4797DB@freenetproject.org> Message-ID: <20080606031841.GC7898@freenetproject.org> * toad at freenetproject.org [2008-06-05 16:48:30]: > Author: toad > Date: 2008-06-05 16:48:30 +0000 (Thu, 05 Jun 2008) > New Revision: 20222 > > Added: > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/ > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > Removed: > trunk/apps/new_installer/res/firefox_profile/extensions/{7DA90D46-1B69-4cc5-9ACE-CB64D8D85B00}/ > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > Modified: > trunk/apps/new_installer/install.xml > trunk/apps/new_installer/res/firefox_profile/localstore.rdf > trunk/apps/new_installer/res/firefox_profile/user.js > Log: > Revert w3 theme. Go back to old theme. > - NOT FREE TO DISTRIBUTE. "Free to use, but contact me for modifications and/or distribution." i.e. requires advance permission from author. > - Inappropriate for Freenet, doesn't look different enough to the default. > This is assuming it works on linux, which we don't know (there is no download for linux). > Heh! what did you do there ? You have reverted r20217, r20216 and r20215 ! I'm fine with you reverting r20216 which shouldn't have reached the repository... but I'd like you to justify about r20215 and r20217. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080606/01c5ae67/attachment.pgp From nextgens at freenetproject.org Fri Jun 6 03:34:48 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Fri, 6 Jun 2008 05:34:48 +0200 Subject: [freenet-dev] [freenet-cvs] r20222 - in trunk/apps/new_installer: . res/firefox_profile res/firefox_profile/extensions res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a } res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a }/chrome In-Reply-To: <20080606031841.GC7898@freenetproject.org> References: <20080605164830.9674D4797DB@freenetproject.org> <20080606031841.GC7898@freenetproject.org> Message-ID: <20080606033447.GD7898@freenetproject.org> * Florent Daigni?re [2008-06-06 05:18:43]: > * toad at freenetproject.org [2008-06-05 16:48:30]: > > > Author: toad > > Date: 2008-06-05 16:48:30 +0000 (Thu, 05 Jun 2008) > > New Revision: 20222 > > > > Added: > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > > Removed: > > trunk/apps/new_installer/res/firefox_profile/extensions/{7DA90D46-1B69-4cc5-9ACE-CB64D8D85B00}/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > > Modified: > > trunk/apps/new_installer/install.xml > > trunk/apps/new_installer/res/firefox_profile/localstore.rdf > > trunk/apps/new_installer/res/firefox_profile/user.js > > Log: > > Revert w3 theme. Go back to old theme. > > - NOT FREE TO DISTRIBUTE. "Free to use, but contact me for modifications and/or distribution." i.e. requires advance permission from author. > > - Inappropriate for Freenet, doesn't look different enough to the default. > > This is assuming it works on linux, which we don't know (there is no download for linux). > > > > Heh! what did you do there ? > > You have reverted r20217, r20216 and r20215 ! > I'm fine with you reverting r20216 which shouldn't have reached the > repository... but I'd like you to justify about r20215 and r20217. And while we are at it, where did you find it's not free to distribute? I have seen no mention anywhere about that! https://addons.mozilla.org/en-US/firefox/addon/6594 http://www.geocities.com/alfredkayser/mozilla/w3v8.htm -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080606/c02b979c/attachment.pgp From j16sdiz+freenet at gmail.com Fri Jun 6 05:11:23 2008 From: j16sdiz+freenet at gmail.com (Daniel Cheng) Date: Fri, 6 Jun 2008 13:11:23 +0800 Subject: [freenet-dev] [freenet-cvs] r20222 - in trunk/apps/new_installer: . res/firefox_profile res/firefox_profile/extensions res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a } res/firefox_profile/extensions/{8e12f188-352c-4476-8198- Message-ID: On Fri, Jun 6, 2008 at 11:34 AM, Florent Daigni?re wrote: > * Florent Daigni?re [2008-06-06 05:18:43]: > >> * toad at freenetproject.org [2008-06-05 16:48:30]: >> >> > Author: toad >> > Date: 2008-06-05 16:48:30 +0000 (Thu, 05 Jun 2008) >> > New Revision: 20222 >> > >> > Added: >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/ >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png >> > Removed: >> > trunk/apps/new_installer/res/firefox_profile/extensions/{7DA90D46-1B69-4cc5-9ACE-CB64D8D85B00}/ >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png >> > Modified: >> > trunk/apps/new_installer/install.xml >> > trunk/apps/new_installer/res/firefox_profile/localstore.rdf >> > trunk/apps/new_installer/res/firefox_profile/user.js >> > Log: >> > Revert w3 theme. Go back to old theme. >> > - NOT FREE TO DISTRIBUTE. "Free to use, but contact me for modifications and/or distribution." i.e. requires advance permission from author. >> > - Inappropriate for Freenet, doesn't look different enough to the default. >> > This is assuming it works on linux, which we don't know (there is no download for linux). >> > >> >> Heh! what did you do there ? >> >> You have reverted r20217, r20216 and r20215 ! >> I'm fine with you reverting r20216 which shouldn't have reached the >> repository... but I'd like you to justify about r20215 and r20217. > > And while we are at it, where did you find it's not free to distribute? > I have seen no mention anywhere about that! Sure you don't read the license when you click "ACCEPT" :) https://addons.mozilla.org/en-US/firefox/addons/policy/0/6594/29665 > https://addons.mozilla.org/en-US/firefox/addon/6594 > http://www.geocities.com/alfredkayser/mozilla/w3v8.htm > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.6 (GNU/Linux) > > iD8DBQFISLBXU/Z/dHFfxtcRApfAAKCXTd3Gd/5pyskqR7KchUYB3M2WLQCg8mt3 > t2CLUsN5bcVFw4Ui/VUpV2g= > =KEV+ > -----END PGP SIGNATURE----- > > _______________________________________________ > Devl mailing list > Devl at freenetproject.org > http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl > From nextgens at freenetproject.org Fri Jun 6 05:31:37 2008 From: nextgens at freenetproject.org (Florent =?iso-8859-1?Q?Daigni=E8re?=) Date: Fri, 6 Jun 2008 07:31:37 +0200 Subject: [freenet-dev] [freenet-cvs] r20222 - in trunk/apps/new_installer: . res/firefox_profile res/firefox_profile/extensions res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a } res/firefox_profile/extensions/{8e12f188-352c-4476-8198- In-Reply-To: References: Message-ID: <20080606053136.GE7898@freenetproject.org> * Daniel Cheng [2008-06-06 13:11:23]: > On Fri, Jun 6, 2008 at 11:34 AM, Florent Daigni?re > wrote: > > * Florent Daigni?re [2008-06-06 05:18:43]: > > > >> * toad at freenetproject.org [2008-06-05 16:48:30]: > >> > >> > Author: toad > >> > Date: 2008-06-05 16:48:30 +0000 (Thu, 05 Jun 2008) > >> > New Revision: 20222 > >> > > >> > Added: > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/ > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > >> > Removed: > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{7DA90D46-1B69-4cc5-9ACE-CB64D8D85B00}/ > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > >> > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > >> > Modified: > >> > trunk/apps/new_installer/install.xml > >> > trunk/apps/new_installer/res/firefox_profile/localstore.rdf > >> > trunk/apps/new_installer/res/firefox_profile/user.js > >> > Log: > >> > Revert w3 theme. Go back to old theme. > >> > - NOT FREE TO DISTRIBUTE. "Free to use, but contact me for modifications and/or distribution." i.e. requires advance permission from author. > >> > - Inappropriate for Freenet, doesn't look different enough to the default. > >> > This is assuming it works on linux, which we don't know (there is no download for linux). > >> > > >> > >> Heh! what did you do there ? > >> > >> You have reverted r20217, r20216 and r20215 ! > >> I'm fine with you reverting r20216 which shouldn't have reached the > >> repository... but I'd like you to justify about r20215 and r20217. > > > > And while we are at it, where did you find it's not free to distribute? > > I have seen no mention anywhere about that! > > > https://addons.mozilla.org/en-US/firefox/addon/6594 > > http://www.geocities.com/alfredkayser/mozilla/w3v8.htm > > Sure you don't read the license when you click "ACCEPT" :) > https://addons.mozilla.org/en-US/firefox/addons/policy/0/6594/29665 > doh! Okay, for some reason I missed it :/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: Digital signature Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080606/a6e73b52/attachment.pgp From toad at amphibian.dyndns.org Fri Jun 6 11:50:22 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Fri, 6 Jun 2008 12:50:22 +0100 Subject: [freenet-dev] =?utf-8?q?=5Bfreenet-cvs=5D_r20222_-_in_trunk/apps/?= =?utf-8?q?new=5Finstaller=3A_=2E=09res/firefox=5Fprofile_res/firefox=5Fpr?= =?utf-8?q?ofile/extensions=09res/firefox=5Fprofile/extensions/=7B8e12f188?= =?utf-8?q?-352c-4476-8198-e9b8f4a4353a=09=7D=09res/firefox=5Fprofile/exte?= =?utf-8?q?nsions/=7B8e12f188-352c-4476-8198-e9b8f4a4353a=09=7D/chrome?= In-Reply-To: <20080606031841.GC7898@freenetproject.org> References: <20080605164830.9674D4797DB@freenetproject.org> <20080606031841.GC7898@freenetproject.org> Message-ID: <200806061250.25701.toad@amphibian.dyndns.org> On Friday 06 June 2008 04:18, Florent Daigni?re wrote: > * toad at freenetproject.org [2008-06-05 16:48:30]: > > > Author: toad > > Date: 2008-06-05 16:48:30 +0000 (Thu, 05 Jun 2008) > > New Revision: 20222 > > > > Added: > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > > Removed: > > trunk/apps/new_installer/res/firefox_profile/extensions/{7DA90D46-1B69-4cc5-9ACE-CB64D8D85B00}/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome.manifest > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/ > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/chrome/tmp.xpi > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/icon.png > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/install.rdf > > trunk/apps/new_installer/res/firefox_profile/extensions/{8e12f188-352c-4476-8198-e9b8f4a4353a}/preview.png > > Modified: > > trunk/apps/new_installer/install.xml > > trunk/apps/new_installer/res/firefox_profile/localstore.rdf > > trunk/apps/new_installer/res/firefox_profile/user.js > > Log: > > Revert w3 theme. Go back to old theme. > > - NOT FREE TO DISTRIBUTE. "Free to use, but contact me for modifications and/or distribution." i.e. requires advance permission from author. > > - Inappropriate for Freenet, doesn't look different enough to the default. > > This is assuming it works on linux, which we don't know (there is no download for linux). > > > > Heh! what did you do there ? > > You have reverted r20217, r20216 and r20215 ! > I'm fine with you reverting r20216 which shouldn't have reached the > repository... but I'd like you to justify about r20215 and r20217. Sorry, would you mind reinstating them? Also do you think we need to rebuild the repository? -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080606/875624e3/attachment.pgp From ian.clarke at gmail.com Fri Jun 6 15:55:18 2008 From: ian.clarke at gmail.com (Ian Clarke) Date: Fri, 6 Jun 2008 10:55:18 -0500 Subject: [freenet-dev] website issue Message-ID: <823242bd0806060855m6819c4d0g3b147b8616c97325@mail.gmail.com> Just looking at the website, the text is right up against the menu to it's left - can someone address this (ideally the person that caused it ;)? Ian. -- Email: ian at uprizer.com Cell: +1 512 422 3588 Skype: sanity From toad at amphibian.dyndns.org Fri Jun 6 23:06:54 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Sat, 7 Jun 2008 00:06:54 +0100 Subject: [freenet-dev] Temporary insert obfuscation: The best of both worlds Message-ID: <200806070006.59846.toad@amphibian.dyndns.org> PROBLEM: If an attacker can identify that each block belongs to the same stream of requests or inserts, he can move towards the originator increasingly rapidly: each request gives him a bearing on the direction (in the keyspace) of the originator, and he can use this information to connect to nodes closer to the originator. Each time he does this, the amount of the stream that he sees increases, and therefore his search accelerates. This attack is easiest on opennet, but it may also be possible on darknet for some attackers (but much slower!). PARTIAL SOLUTION: We have a partial solution in that we don't insert the top block, or generate the key, until after we have inserted all the data below it. However, in many instances the data will be guessable (or guessable to within some computible entropy e.g. all but a few bytes) and therefore the attacker can still identify the blocks. BAD SOLUTION: One proposed solution (for inserts) is to insert each splitfile encrypted with a random key. This would help a lot with this attack (for inserts), but it would cause much more data duplication, entirely losing the benefits of convergent encryption (CHKs). To what degree convergent encryption is useful is an open question, but there is a better way... NEW PROPOSED SOLUTION: When Alice inserts a file, her node runs FEC encoding as usual. Then a random obfuscation key is chosen, and each block is encoded, both with the random obfuscation key, and with the normal CHK convergent encryption. Alice's node inserts only the obfuscated blocks, but it computes the CHKs for the non-obfuscated blocks if they were inserted. The top block includes pointers to the top level of the splitfile for both the obfuscated and non-obfuscated versions, but only the obfuscated version has been inserted *by Alice*. Alice announces the key on a public forum. Bob (hopefully there will be many Bob's) starts to download it. For each block in the splitfile, Bob's node tries the non-obfuscated block first, maybe giving it a 3-try head start. Then when this fails it tries the obfuscated block. When each obfuscated block is fetched, there is a chance that the non-obfuscated version will be inserted by Bob's node. Thus, Alice is protected by Bob. Most likely Alice is in the greater danger: generally you want to go for the source of the data. The attacker will then only gain a small amount of information from a splitfile insert: even though he can identify the blocks in retrospect, he can't move towards the insertor during the insert, so he gathers much less information. It will probably take more than one splitfile insert to trace Alice. Of course, splitfile inserts aren't the only thing that gives him bearings on Alice's location, her FMS posts etc (e.g. announcing her files) will also betray her in sufficient quantity. It would be good to have some sort of guesstimate that you have to change identity every X messages or inserts to be reasonably safe... Sadly protecting requestors in this way is impossible afaics (although in the long term, premix routing and/or tunnels will help). And in the long run we don't fetch the obfuscated data, only the non-obfuscated data, which will collide for multiple inserts of the same data, so we don't waste any space. Obviously this whole mechanism would be optional so that you can turn it off for a slightly faster / less demanding insert (but with less security). Problems: ** Whether to include the non-obfuscated blocks in the splitfile metadata. ** Inserting the non-obfuscated blocks when we get the obfuscated ones, and trying the non-obfuscated ones before the obfuscated ones, is intended to result in the non-obfuscated blocks being preferred, and the obfuscated ones falling out of the network. We have two options: 1. We could have a parallel metadata structure, with only the obfuscated keys in the obfuscated metadata, and only the non-obfuscated keys in the non-obfuscated metadata, and the top block referring to both, but Alice only inserting the obfuscated metadata. CON: - A requestor will not know the obfuscated blocks' keys until he has decoded them. Thus, he cannot request them, until some requestor manages to download the entire (obfuscated) file and insert the non-obfuscated metadata. 2. We could include the keys of both the obfuscated and non-obfuscated blocks in the obfuscated metadata. PRO: - We can immediately fetch the non-obfuscated blocks from day one. So we can further propagate the blocks inserted by our fellow requestors, and not propagate the obfuscated blocks, maximising efficiency and resulting in faster downloads when a file is new. - Also, if the same file has been inserted previously, we can immediately pick up the blocks inserted for the other file. CON: - The obfuscated metadata will be twice the size because twice the number of keys. And it will not collide, because it is obfuscated. MITIGATION: - However, we can have non-obfuscated metadata in parallel. This could be inserted by any requestor, and would be the same size as splitfile metadata is now. Bob's node would request both the obfuscated and the non-obfuscated metadata initially, but if the non-obfuscated metadata was found, there would be no need to fetch the lower layers of the obfuscated metadata unless they were needed i.e. problems were encountered fetching the data. IMHO we should implement the second option, with the mitigation so that long-term very few blocks are fetched from the obfuscated data, and it falls off the network. ** Possible optimisation: Timestamps ** We could include a couple of timestamps in the top level metadata: - Before time T1, fetch the non-obfuscated data as often as the obfuscated data. - After time T2, don't fetch the non-obfuscated data at all (or only if really desperate). Obviously this would have to be optional, and the original time of insertion would have to be obfuscated. And this will only work if we are sure that our data will be downloaded by several others shortly after we announce it. -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available Url : http://emu.freenetproject.org/pipermail/devl/attachments/20080607/004a80d9/attachment.pgp From toad at amphibian.dyndns.org Fri Jun 6 23:46:54 2008 From: toad at amphibian.dyndns.org (Matthew Toseland) Date: Sat, 7 Jun 2008 00:46:54 +0100 Subject: [freenet-dev] Temporary insert obfuscation: The best of both worlds In-Reply-To: <200806070006.59846.toad@amphibian.dyndns.org> References: <200806070006.59846.toad@amphibian.dyndns.org> Message-ID: <200806070046.59876.toad@amphibian.dyndns.org> On Saturday 07 June 2008 00:06, Matthew Toseland wrote: > PROBLEM: > If an attacker can identify that each block belongs to the same stream of > requests or inserts, he can move towards the originator increasingly rapidly: > each request gives him a bearing on the direction (in the keyspace) of the > originator, and he can use this information to connect to nodes closer to the > originator. Each time he does this, the amount of the stream that he sees > incre