[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Philosophical question



Larry G. Gariepy Jr. wrote:
> 
> Hi all,
>     I guess this doesn't pertain specifically to Debian, but I have a
> philosophical question about software upgrades, and I just thought I would take
> a poll to see what Joe Debian User (or Jane Debian-User) thinks...
> 
>     So I bought a new machine in October, and from the outset I wanted to run
> Linux on it.  After asking a friend what brand to use, he suggested Debian.  In
> fact, we basically downloaded his setup onto my machine.  So now that I
> basically have the system up and running the way I want it (modulo getting the
> printer pass-through feature working on the ZIP drive), I thought I would ask
> how often I should look to upgrade things like the kernel and libc libraries
> and such.  I am not shy about trying new things, but once I have put a lot of
> effort into something and have it running nicely, I tend to be sluggish about
> rocking the boat, and upgrading to keep up with the times...(when I was a kid,
> we didn't even HAVE xemacs, we used all those CNTRL commands... :)  )
> 
> I hope I don't sound heretical:  I imagine that a lot of people like Linux for
> the fact that there is something new to try every week or two. :) (or more
> often?)  But practically, how often do people think it is worth the trouble to
> upgrade major software components?  Every six months?  Every year?
> 
> Thanks in advance for your (humble) opinions :)
> Larry Gariepy
> gariepy@dartmouth.edu
> 
> Disclaimer:  The views expressed in my opinions are not necessarily my own.
> 
> --
> TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
> debian-user-request@lists.debian.org . Trouble?
> e-mail to templin@bucknell.edu .

There is no correct answer to this question.  There are wrong ones
however.  Installing packages (updates) simply because they're newer is
not the right strategy, nor is installing simply on the basis of new
features.  The steps I normally take go something like :

1. Will it be as secure or more secure than what I am using now?
2. Will it increase performance?
3. Does it expand the feature list of the current software?
4. Is this a major version number update?

Using these 4 rules I can usually decide whether or not to upgrade. 
Normally if question one proves to be false I stop right there.  In some
cases there really is no question one, and so it becomes a performance
and feature issue.  Number four is tricky, I've found that the best
strategy is to stay away from major version changes.  They tend to mark
the period for bug testing, and usually most problems and glitches arise
here.  Going with the previous version in this case tends to be a good
thing.

Hope that helps somewhat... sorry that I couldn't give you a time frame,
but these rules generally help me to do the right thing.


--
TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to
debian-user-request@lists.debian.org . 
Trouble?  e-mail to templin@bucknell.edu .


Reply to: