On Wednesday 21 September 2005 19:53, Siddhartha Basu wrote: > Hi, > > On 9/20/05, A. Mani <[EMAIL PROTECTED]> wrote: > > On Tuesday 20 September 2005 20:02, Sharninder wrote: > > > > > 3. Documentation, whitepapers and articles > > > > > > > > projects > > > > > > > > > are always needed. > > > > > > This reminds me that ilug-cal had started an > > > inititative of this sort a few years back. I > > > remember 'cause I also wrote a small'ish howto for > > > that. Is that project still live ? What came off it ? > > > > > > -- > > > Sharninder > > > > There are lots of material coming up on the net in the linux world. If we > > need to do something, then there must be a real need for it. If you take > > all the factors there, then it becomes an optimisation problem (solvable > > usually). We must in particular pay more attention to local problems. > > Can you specify some local problem that might require documentation.
1. A basic inference about security in general is that one must at least use the latest s/w. But when it comes to relatively slower PCs (which are more common locally), a perfect set of specifications for linux distros is of much relevance. 2. Reporting / Documentation on OSS s/w testing for different categories of users. 3. We can build a collection of links to other resources in the area...wiki-style. 4. Specific Howtos for administrators in key fields incl. schools. Often the problem with administrators is "he will come and do it".... 5. Comparisons between s/w. 6. H/w compatibilty and evaluation. > >It certainly helps > > if people identify issues (on the list) which require more detailed > > anaysis. > > Just an idea. One pattern i observed is the interest in testing of > various distros by our users. What happens is that sometimes though a > solution is posted/suggested it really never get solved because it > might not be applicable to that particular distro. There were also > times when an user with a probable solution doesn't no how to handle > the situation with another distro. Sometimes there are polymorphism of > the similar problems. Not to mention, people are sometime even > clueless about properly defining their problems. Following on that > thought, we could write a collaborative short howto about different > popular distros by their respective distro specialist in our list. It > should aim to provide a short summary on what kind of audiences it is > targeted (newbie, intermedia, moderately experienced, guru), what kind > of job it is better suited for doing(desktop, multimedia,server, > developmental, educational, enterprise) and why, how basic > installation and configuration are handled(installation, init system, > X-server configuration, system administration, multimedia setup, web > surfing/e-mail), pointers for distro specific documentations. It > should help people to select their compatible distro and reduce their > downstream problems. Of course, in the mailing list it will help us to > encounter some repetitive problems, we could properly organize our > thought,time and energy to do some more constructive projects. > Certainly. Bakshi did a lot of testing with wms, recently and it will help if it is turned into a nice report. > > Then if some others can, then we can have nice articles, documentations > > etc on our site. Then again we might identify such needs through a > > thread. That is how it stands. > > We do have a "Newbie FAQ" project in our web site which doesn't seems > to have a lot of activity recently. I am ready to spend at least (2-4 > hrs/week) to keep that going and more as it will require. If somebody > is already assigned with that project let me have some pointers to > start again. We definitely need some more volunteers with more > questions and answers. > So, who is looking after that project, let me know. who ? > > > Now performance evaluation in this kind of scenario must necessarily be > > dynamic. It requires quite some effort to quantify it (an interesting > > topic if you want to use such things to build examples for your theories, > > like I sometimes do) in some sense. > > It is not quite clear what exactly you meant to say here. Could you > provide some detail and/or some examples. Take the case of localisation as far as language is concerned. The basic tools available for this are only available in part, but is available to a relative maximum level (from the global perspective). This is working fairly well with lots of groups involved across the country. Now this means ilug-cal can bother less on the issue of maintaining howtos, articles on the issue ...but links maybe deemed to be compulsory. We may prefer to deal with this by restricting to a set of relevant packages. Now I was actually thinking about modelling the situation over time for evaluation. This can be done in various ways depending on the amount of information available. For example via rule-extraction methods admitting vagueness and uncertainity. > > > But what is more important is that such efforts should be persistently > > motivated. > > This thread is definitely a step towards that effort. Lets keep it on. > > -siddhartha > > > A. Mani > > Member, Cal. Math. Soc > > A. Mani Member, Cal. Math. Soc -- To unsubscribe, send mail to [EMAIL PROTECTED] with the body "unsubscribe ilug-cal" and an empty subject line. FAQ: http://www.ilug-cal.org/node.php?id=3
