Home Messages Index
[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index

Re: [News] Costs of Linux Development, Scale Benchmarks (My O/S is Bigger Than Your O/S)

  • Subject: Re: [News] Costs of Linux Development, Scale Benchmarks (My O/S is Bigger Than Your O/S)
  • From: Mark Kent <mark.kent@xxxxxxxxxxx>
  • Date: Wed, 13 Sep 2006 07:03:57 +0100
  • Newsgroups: comp.os.linux.advocacy
  • References: <2735166.Ax6LoDprMA@schestowitz.com> <slrnegdt8o.t0a.sorceror@localhost.localdomain>
  • User-agent: slrn/0.9.7.4 (Linux)
  • Xref: news.mcc.ac.uk comp.os.linux.advocacy:1153857
begin  oe_protect.scr 
Ray Ingles <sorceror@xxxxxxxxxxxxxxxxxxxxx> espoused:
> On 2006-09-12, Roy Schestowitz <newsgroups@xxxxxxxxxxxxxxx> wrote:
>>| Operating system                           Lines of code
>>| Microsoft Windows 2000 (February of 2000)     29.000.000
>>| Microsoft Windows XP (2002)                   40.000.000
>>| Debian 3,0 (July of 2002)                    105.000.000
>>| Debian 3,1 (June of 2005)                    229.500.000  
> 
>> What's noteworthy is that the fairly small Windows is in a state of utter
>> mess, which requires over 20 million LOC's to be (re-)written. On the
>> contrary, Debian is renowned for stability, reliability and security,
>> despite its large scale, function, and flexibility.
> 
>  The actual study (in English) is summarized here:
> 
>  http://www.elrincondetolito.com/archivos/Measuring%20Libre%20Software%20Using
> %20Debian%203.1%20Sarge%20as%20a%20case%20study.pdf
> 
>  It's not really fair to Microsoft there, however. I'd bet that if you
> included all the official MS-branded programs out there (not the
> third-party stuff), you'd have a comparable lines-of-code count. Of
> course, the above count *only* includes the "main" portion of Debian,
> not the "contrib" or "non-free" stuff.
> 
>  But what's interesting is that the Debian code was developed, and is
> available, at a *far* lower cost than the comparable Microsoft
> collection.
> 
>  The paper's authors note:
> 
>  "However, there are some aspects that put into doubt the future
> sustainability of this progress. For instance, mean package size
> is showing an unstable behaviour, probably due to the number
> of packages growing faster than the number of maintainers.
> Nor can we forget that we have had to wait almost three years
> for a new stable release and that the release date has been seri-
> ously delayed on several occasions."
> 
>  I think they are a trifle alarmist about their quibble about package
> size, particularly since they are talking about an increase from 23,000
> lines to 26,600 lines. But organizing that many packages, even when the
> work can be distributed, *is* a daunting task.
> 

To be honest, I'd say that debian should be running at its limits, in
particular, there should be more packages than there are maintainers to
go around.  This means that the best possible debian is coming out of
this process, and, of course, if anyone thinks something is missing,
they can become a debian developer and maintain it themselves, or
indeed, maintain a local version, or pay someone else to do it.

There will be a constant need to recruit more people into the
maintenance process, /but/ as it's possible to do real PC programming
with virtually no investment now, I can't see a shortage of newbies
coming along - FOSS has ended that - anyone can learn.

-- 
| Mark Kent   --   mark at ellandroad dot demon dot co dot uk  |
	"... freedom ... is a worship word..."
	"It is our worship word too."
		-- Cloud William and Kirk, "The Omega Glory", stardate unknown

[Date Prev][Date Next][Thread Prev][Thread Next]
Author IndexDate IndexThread Index