Jump to content

Eduard Nicodei

  • Content Count

  • Joined

  • Last visited

Everything posted by Eduard Nicodei

  1. Hello, I'm curious to find the repository where dev-python/cryptography-0.9-r1000 is pulled from. I checked the log but unfortunately there are only references to the staging repo (which apparently is not public) # git log cryptography-0.9-r1000.ebuild commit 8f623b52b94bb842ae439f4c368f0080c75fe6d7 Author: Daniel Robbins <drobbins@funtoo.org> Date: Wed May 27 16:24:45 2015 +0000 merged from funtoo-staging merged: funtoo-staging: 08c2e2031707b845142d932cdcc5d3def1633e76 On this note, would be possible revert to the old-style of commits where individual SHAs were shown for each repository? So far, dev-python/cryptography-0.9-r1000 it doesn't seem to be in "funtoo-overlay" or vanilla Gentoo-port tree . I've also checked the Repositories but it's not there either. Searching on Google, I could only find references to "zugaina", but I believe we don't have that overlay in Funtoo right? EDIT: actually I've found it by manually going through each of the repositories, apparently it's from gentoo-progress. Still, is there anyway to find this without going through all the repositories in a browser? Thanks, Eduard
  2. I'm going out of a limb here, but reading through the wiki: Could somebody maybe provide a bit more motivation for having such a goal? On first glance it seems like the request for a lot of duplicated effort. The Gentoo wiki has a lot of documentation and the Arch Linux Wiki is probably one of the best resources for packages of all kind, not to mention the man page (although it is true that some man pages completely lack any depth). Why am I asking this? Let's take an recent page that was added to the wiki: http://www.funtoo.org/Package:Tmux Although max kudos to the person who wrote this, isn't all of the information present on that page already available elsewhere? I completely agree that writing documentation / help is possibly the most important (non code related) contributions to be made to opensource but is such a massive duplication of work necessary? Shouldn't the Funtoo wiki restrict itself to Funtoo-specific changes in the packages? Now I might be horribly wrong here and I apologize for making ignorant claims, but shouldn't there be some clear "Why do we want to do this / Motivation" section in addition to the "Goal" one in http://www.funtoo.org/Adding_an_Ebuild_to_the_Wiki ?
  3. You can keep using gentoo-sources if you like. AFAIK the difference in kernel source packages is simply the patches applied to them (somewhat distribution specifc fixes) In any case, you can definitely boot into either a Funtoo CD or any live CD and copy its configuration to your own /usr/src/linux directory. Personally, I've based my kernel on a Ubuntu 12.04 config. It provided everything and more. I then started removing stuff that I didn't want. I also configured some drivers such as SATA ones to be built into the kernel instead of as a module. This way I don't have to use initramfs. You can get the live CD config either from something like /boot/, or more likely, as sputnik pointed out, from /proc/config. Simply copy it in /usr/src/linux, chroot and make.
  4. tl;dr: want my voice heard through more surveys. Hi all, I was reading about the new Kickstarter campain of Operating System U. Its basic premise is that it should provide a familiar and well known interface to users. Indeed, people don't like it when things change. Leaving that aside, it got me thinking: "why do these changes, that make the people unhappy, happen in the first place"? I think there could be a few examples of "new technologies that people don't like", think for example systemd, Gnome3, Windows 8, each time Youtube's interface gets upgraded or one's favourite API becomes deprecated. If one looks past the objective performance/usability changes, the reason some people don't like these programs might be simply because they change how things are done. So why does that happen? The only reason I could come up with is how we do development: developers develop for some time and then release to the public. But if the public's response is one of anger and frustration, then the developers start arguing for why their release is better (faster, smaller, less bugs, more features etc). But notice how now we're arguing about two different things? One camp has the problem that they don't understand the new API / interface. The other camp has the problem that the first camp says the software is s**t. But these are different. So my first question is: do you think the above is something that happens? - for all I know I might be completely out of sync with reality so better take things one step at a time... :) If it indeed is a problem, how can it be solved? Again, one solution I could come up with is having surveys. Yes, surveys. Generally every respectable hotel you go to will have a feedback form. Every university course will have a feedback form at the end. Commercial websites will ask you to do surveys when visiting their website so they can "improve user experience". Could this be applied to opensource projects as well? Other commercial products invest in Human-Computer-Interaction (HCI) research, again to improve their product - why? Because developers aren't very good at making HCI-related decisions. What would happen if opensource projects had an easy way for regular users to send feedback? What if we used something like Ubuntu's brainstorming, where users could post improvements and vote on them? Could this be done right though? Could new or uninformed users make bad suggestions? Note I'm not talking about bugs (broken things), but rather improvements if only extremely slightly. We of course have ways of users to suggest improvements. But why not ask them? Why not ask users what they want, or what they think of feature X, or what they think of feature X given that feature X is awesome because such-and-such? Why do this? Because it will make software better for the user. It is debatable whether it will make it better for its developer, however I think that people are first users and then developers. By making software more user-friendly we encourage users to use our software and chances are some of them will contribute back, thereby further improving it. Now as I mentioned, this is something that popped randomly into my head so I might be horribly wrong. If not, may I propose that we have polls with possible new features that could be added to Funtoo forums? One such poll might even be if such polls are a good idea :) Also, thanks for reading!
  5. Hi Don James, I had the same problem a while back. I remember doing a Google search which probably led me to http://forums.gentoo.org/viewtopic-p-7465170.html My solution was to re-emerge python: emerge python:2.7 python Before doing that please note the output of eselect python list It might be that your default python interpreter is 3.* in which case it might be also worth trying to switch to python-2.7 (using eselect python set NUM) and then trying to reemerge chromium again. (Also, as salazaraj2 pointed out, make sure you actually have the option enabled in the kernel) Hope this helps!
  6. I had the same problem (last sync was Jul 26). Out of curiosity, does anybody know why this happens? It seems odd that a git repository gets corrupted out of the blue. One of the design principles behind git is that it ensured data consistency. Is this therefore caused by improper handling of the repo by github? Or is it caused by something like a complex history rewrite (git push --force)? I'm asking because the comments on the bugtracker don't say anything about the underlying cause of this bug. Thanks in advance! edit: to fix, I first did a git status on /usr/portage, then simply copied /usr/portage/distfiles and /usr/portage/packages to a safe place, rm -rf /usr/portage/.git then did a emerge --sync and copied back distfiles and packages (basically plans D-F but also saving my package cache).
  7. Thanks. Now tracking on http://bugs.funtoo.org/browse/FL-1278
  8. Hello all, I noticed we have a separate ebuild for xchat in our overlay. The cause seems to be FL-295 (funtoo-overlay: git log -v net-irc/xchat/), which is related to Gentoo bug 420291 However, Gentoo bug resolution is "Won't fix". This seems to be because Gentoo removed xchat from their tree in favour of hexchat: bug 460038. Should we reflect their changes in our tree as well? I'm new to Funtoo development and would like to get involved. I would be happy to create a JIRA entry and resolve the issue myself. I am asking here first because I want to double check this is the correct course of action. Many Thanks! Eduard
  • Create New...