NM fails to connect when booting ?? -[SOLVED]
Bill Davidsen
davidsen at tmr.com
Wed Mar 18 22:11:39 UTC 2009
Mark Haney wrote:
> Bill Davidsen wrote:
>
>> I had actually hoped that the "profiles" stuff might let me have
>> multiple configurations which work at multiple locations. I have three
>> locations needing credentials, and I'd like to have a "Starbucks"
>> configuration which found whatever was available. Doesn't seem to work
>> that way. :-( Thankfully I know how to write scripts and use iwconfig...
>
> Yeah, it's a saving grace for those of us who have given up on NM.
>>> NM sucks, everyone knows it, so let's can it and move to something that
>>> doesn't.
>>>
>> The reason people think it sucks is that the documentation is missing,
>> inadequate, or wrong. And when I mentioned this someone told me that
>> writing documentation is not a good use of developer time, and that's
>> hogwash. I've done FOSS and commercial development for over thirty
>> years, and on commercial software there was always a description before
>> the code was written, and someone writing documentation which was
>> checked by QA, or for something I was giving away, I wrote my own
>> because I wanted people to love and use my software. We used to call the
>> NM approach "if it was hard to code it should be hard to use."
>
> Actually, the documentation issue isn't why I think it sucks. But you
> are correct, no/poor/incorrect documentation is BAD. Very bad. I've
> been a sysadmin for 12 years now professionally and done a great deal of
> programming, I document /everything/ even in the smallest script. It
> makes life MUCH easier.
>
> The reason I think it sucks is because it's so erratically unstable and
> buggy. I'm all for open source, but implementing a half-baked app like
> this for something as important as the network is just foolish. This
> also coincides with bad documentation, I've not looked at the code, but
> if they don't document the app, is the code undocumented as well? To
> me, it seems that that is the case based on the unstable nature of the app.
>
I was looking for an essay I once had, called "Comments are Documentation, Too."
It undoubtedly came from my youth, when structured programming was still used.
In my search I found a link to "Documentation: Give it up; it won't happen"[1]
and "FLOSS Manuals sprints to build quality free documentation."[2] I hate to
say it, but "quality-free documentation" comes closer to the case here.
> The decision to use this is still beyond me...
>
>> I would would be ashamed if something I wrote was a constant topic on
>> this list, had it's own list on how to avoid it, even had people selling
>> bloody tee shirts which show "NetworkManager" in a circle with a slash
>> through it, and generated opinions like the "everyone knows it" above.
>> Actually I exaggerate, the site was selling iron-on sheets for $4, you
>> had to provide your own shirt, but there must have been some market. I
>> do believe that a scan of this list indicates it's not just a few people
>> frustrated by the lack of documentation.
>>
To the person who commented that I "probably never really wrote any ____ing
documentation at all," a number of my articles have been published, you may be
able to find the ones I wrote for _SysAdmin_ magazine a few years ago still online.
[1] http://www.linuxjournal.com/node/1005775
[2] http://linux.com/feature/155205
--
Bill Davidsen <davidsen at tmr.com>
"We have more to fear from the bungling of the incompetent than from
the machinations of the wicked." - from Slashdot
More information about the fedora-list
mailing list