On Sat October 8 2005 11:48, elefino wrote:

On Friday 07 October 2005 18:25, James Ogley wrote:

...

...

What have I done wrong. I want firefox to come up when I click a link in evolution. I do not want or like epiphany .

What have you done wrong? You've monkeyed around without knowing what you're doing.

Change your GNOME web browser to Firefox. Sorted.

How does one find out what one is doing, without monkeying around?

Information about Linux stuff, on the web, doesn't seem to be tagged for freshness, with expiry dates and applicability notes. Just reading everything to be found on a topic, before doing any monkeying has never done me much good... and I'm one of those people with a need-to-read (I even re-read the label on the cereal box if there's nothing else handy...). I've found that wide reading means reading of lots of contradictory and outdated stuff, along with stuff that might be valuable. I do it, but these days only to get the general flavor of the area that currently has me stymied, not for specific steps, since following steps from two (or ten) contradictory HowTos or extracted from 2002 Debian mailing lists (hey, if that's what Google puts up, that's what I read) is a recipe for trouble as well.

Whenever I start anything I haven't done before, I read man pages and look for HowTos, but I invariably get into trouble because - without experience - the multiple ways to do anything in Linux don't sort themselves into neat categories. Later, when somebody on the list helps me to dig myself out, it usually involves backtracking, uninstalling and deleting 47 things that I didn't need to do, but which conflict with the one path that I've (or the coach) settled on.

I can short-circuit the process sometimes, by just jumping in, getting in trouble, and then getting help to extract myself. That approach at least saves me from reading dozens or hundreds of pages that don't apply. :-)

On average, it takes me many weeks or months to sort out any installation/configuration that doesn't accidentally go perfectly the first time. Thus, if I start trying to do things shortly after I install the latest SuSE, then I may actually have time to get to a resolution on some little project. If I start a couple or three months after the last SuSE release, then my shambles remains and my motivation wanes (because there's a new release coming up that might fix it...) until I install the next SuSE, and likely it gets fixed automatically... or else it's now a completely different problem that prevents whatever I wanted from working.

For example, having installed 10.0 two nights ago, it will now take me at least a couple of hours to get DVD-player stuff working (including finding my decss stuff, finding un-pre-broken rpm* of xine that's intended for 10.0 and not some earlier SuSE, reading through old SLE posts to get the procedure straight, agonizing over not finding the rpms where they were the last time - or finding them only for 9.3 -, etc., whereas you probably have a script that does it in ten minutes, including download time. But it took me three or four months with SuSE 9.0, a couple of years ago, because of all the wrong paths I took (including two months to get apt-get working, because that was how the most consistent and encouraging thread in SLE explained it).

Different world-view. Different talents. Different use for the computer. I know it's not pretty, but that's what you are dealing with, here, and I'm not alone. When Linux went beyond pure command-line geek toy, people like me came out of the woodwork, and most of us don't have the sense to go away. :-)

Kevin (* I hold out for rpms because I don't want to undermine the rpm database and YaST's/YOU's usefulness)

Thank you for taking the time to address this issue. I have also found that the Linux documentation issue is lacking. I believe it is because Linux came from unix, and the unix crowd has a sub-set of gurus who (a) knew what they were doing, and (b) wanted to maintain their position if intellectual dominance. This is fine for what unix used to be. In the mini-computer world of business, after all, these computers sold for 3000 to 15000 dollars, the 'users' had no interest or need for the inner workings, they were 'appliance operators'. Now, Linux is entering the mass market world, and in that world there is a specific percentage of users who want/need to know the inner workings (such as me). Your 'date' issue is on target. Consider dosemu. The doc discusses an old version that does not install/work the same as the current versions. There should be a standard of documentation. They should pass the fog test (a program the determines the level of education needed to understand a document) for about the 8th grade. When I ran a fog test against our training manual for my company, I found that that you needed a 2nd year college ed to understand it. Sounds nice, but that means that we were not communicating with our staff. Same for linux. Tutorial means step by step. Introduction means that words are not used (verbs that equate to linux programs (rpm, etc.) that are not defined. Documents are dated when created and when updated. The version of the program being documented is referenced. I could go on, but I am not a professional writer, and that is what is needed to create a documentation foundation. I remember the Radio Shack Model I, my first computer, in 1977. The author of the manual, David Lein?, had a Phd. (doctor of thinkology). The manual received aclaim because it targeted the audience so well, and probably sold more computers. I think this is a growing pain of the open source concept. -- John R. Sowden AMERICAN SENTRY SYSTEMS, INC. Residential & Commercial Alarm Service UL Listed Central Station Serving the San Francisco Bay Area Since 1967 mail@americansentry.net www.americansentry.net