Re: Has The performance been forgotten?



-->> Also, please read other posts I send
simultaneously.

My friend,

If a project is deflecting from the right path, It may
be necessary to stop its development and revise some
past steps.

I regret I'm not a programmer and cannot participate
in development, and sometimes cannot understand how
much difficult can be offering a fast, beautiful, and
bug-free software.

May be a question about slowness of SuSE lead to one
of its components e.g. KDE; and that question about
KDE lead to other projects like QT; Similarly, many
projects may be known responsible for a certain issue,
but all in uncertainty, and all may deny! If such a
lack of co-ordination can be found in the linux
community, solving a performance problem becomes
disappointing; However, I hope this not to be true.

If hopefully, the above deduction proves flase, one
can ask developers to work seriously on some projects
with the focus on performance: a parameter that is one
of the most important criteria which unfortunately,
only very skilled programmers pay attention to it.

The performance issue here is not around %2 or %10: We
have a great unreal slowness that requires severe
notice; When your N hours spent on it saves N*10000
hours of user's times, isn't it worth doing a tedious
work?

Just at the time that open-source products are
reaching to a usable, favourable, and likable point,
adding a series of unnecessary cumbersome features
make it so ugly and awkward that deservers laughing at
it again.

Isn't it a good impression to say that "THE FASTEST
LINUX" when introducing a new version or distribution?

I hope developers hear this little user's voice and
promise to examine the case.

Thank you for your notice,
Bahram Alinezhad,
Tehran, Iran.

-->> Also, please read other posts I send
simultaneously.

---------------------------------------------------
"Jeffrey L. Taylor" (suse austinblues dyndns org)
wrote:
---------------------------------------------------

Developers often have the latest and fastest machines.
 It isn't that they ignore performance hits.  There
aren't any on the machines they use.  Volunteer
developers (i.e., non-paid) get no brownie points for
improving performance.  They do get brownie points for
features.  They get flak for really bad performance. 
Flak can be ignored, discounted, brushed off, etc.  If
you want performance, reward it with money, hardware,
public praise, etc.  Performance tuning is hard,
tedious work that mostly yields little reward.  (How
many brownie points are you going to get for a 2%
improvement, even if it took a week or two to find? 
And performance tuning is a game of diminishing
returns, i.e., if the first week of tuning may get you
a 10% improvement, the second will probably yield no
more than half that.)

Jeffrey



		
_______________________________
Do you Yahoo!?
Express yourself with Y! Messenger! Free. Download now. 
http://messenger.yahoo.com



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