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

Re: [nathan@dnase.hpw.pri.bms.com (Nathan O. Siemers)] Humble Request Re: New release over due



Chris Waters <xtifr@dsp.net> writes:

> Newest features and stablity are usually incompatible goals.  Also,
> most businesses find upgrades both expensive and time-consuming.  A
> release a year is FAR TOO OFTEN from one perspective, and nowhere near
> often enough from another.  Where to find middle ground?
> 
> For everyone whining that Debian doesn't release often enough, there
> are three people switching their businesses to Debian because it's
> solid and reliable and stable and doesn't change all the time, unlike
> most other distros.  I think we're doing ok.

	I agree in principle with some of the above.  Debian systems
	are stable as we all know. However, releasing more frequent
	updates with more modest agendas does not force any system
	administrator to upgrade more often - it only adds flexibility
	for them to add new features when necessary.  Also, the less
	you change at once in the overall debian architecture
	(changing package structures, package methodology, etc), the
	easier debugging the organic whole will be.

	My original point was that potato has gotten delayed beyond
	it's (my?) anticipated release because developers tried to do
	*too much* in a single release.  It was a call to modesty, not
	a call from a feature hog.  My job involves running a
	computational infrastructure group within a fortune 50
	company.  We have around 60 linux systems here, including two
	compute farms.  Where is my middle ground? I cannot be on the
	bleeding edge, but slink is looking *very* old at the moment.
	Our group is research oriented, so we are less conservative
	than many, but I do represent enterprise computing to some
	extent :)
	
	Someday these at work machines will be running debian
	gnu/linux, like my 3 systems at home.

	nathan


> 
> If you really need the latest and greatest, use potato now; we all do.
> (Every Debian developer is pretty much forced to run unstable, which
> ensures a certain level of stability.)



-- 
N a t h a n   O .  S i e m e r s
Bioinformatics
Division of Applied Genomics
Bristol-Myers Squibb Pharmaceutical Research Institute
Hopewell Building 3B, P.O. Box 5400, Princeton, NJ 08543-5400
609 818-6568
siemersn@bms.com


Reply to: