[ Date Index ] [ Thread Index ] [ <= Previous by date / thread ] [ Next by date / thread => ]
On Sat, 27 Oct 2007 21:55:59 +0100 tom@xxxxxxxxxx wrote: > Many people say that OSS is about scratching an itch. True - but there are a lot of areas to cover. Scratching an itch is all about getting involved in a specific area and there are so many options available, no two programmers end up with exactly the same itch. I can only suggest that people start with what they know and find something that they feel is within their grasp. > I am sure that there are > many of use here who would like to be able to do this. Not major changes but > maybe changing certain parts of the UI etc.. OK, that limits things a little, but not much. > For those of us who are not > experienced in programming but would like to try and make a change it can be > very hard. I personally find I just fumble around trying to would out how to > change something that should really be simple. It's never more than a SMoP. [Simple Matter of Programming] The problem comes in moving from spotting the problem to understanding the solution. I'm all for helping out with this complex process but there do have to be limits on what can be done. > If the people with the simple English language instructions could meet with > people better versed in a particular language then some helpful changes could > be achieved and hopefully the first person would have a better idea of where > to start on the next patch. Yes, but there are several problems here: 1. The people reporting the bug should not be expected to limit their reports to specific languages. Bugs just don't work that way. 2. There are too many languages, libraries and subdivisions for those who do programming in the LUG to cover more than a small proportion of the packages involved in the bug reports. 3. Add in distribution differences and the process gets even harder. Even between Ubuntu and Debian there are significant issues that mean a particular bug could be unreproducible. That is why bug tracking systems exist - they are supposed to allow people who don't understand programming to get appropriate answers from those who do, by always linking the reporter with the one person who should know how to fix it - the package maintainer or upstream. Asking a random programmer to debug a package s/he has never seen before is seldom useful, especially if the package is not one that the programmer has any particular interest in fixing. > Would people from either side be interested in a meeting where we create some > simple patches, have them explained in detail and then go through the process > if submitting them? Only if the limits of such a meeting are precisely defined. e.g. I'm not volunteering to create patches against packages outside Debian and even within Debian, I only have sufficient experience to look at C or Perl and only within Gnome or userspace command-line utilities. I'm up for a general discussion but it will be hard matching bug reporters with patch creators for the same package within what is still quite a small group. -- Neil Williams ============= http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/
Attachment:
pgpR05lmrdZd5.pgp
Description: PGP signature
-- The Mailing List for the Devon & Cornwall LUG http://mailman.dclug.org.uk/listinfo/list FAQ: http://www.dcglug.org.uk/linux_adm/list-faq.html