Delay? Looks bad for Fedora

[Mike A. Harris]

On Mon, 3 Nov 2003, Mark Hutchinson wrote:

>A bit of a rant, but some serious point on Fedora.

Not really.

>What is the delay due to on the release?  

Read the list archives, or the previous mails in your mail 
folder.


>I do not understand what this means: Delayed due to respin for several days, 
>will be available as soon as possible.....

Read the list archives, or the previous mails in your mail 
folder.


>This may have been posted before, but I missed it.

Read the list archives, or the previous mails in your mail 
folder.


>This is not a good sign for those waiting to see if fedora will
>still be a professional distro.

Oh really?  Kindof funny.  Other people consider the sign of Red 
Hat intentionally delaying a release in order to fix some very 
important last minute issues to be a loud sign that Fedora is a 
serious distribution, and not just some rag tag thrown over the 
fence.

There are always last minute surprises when creating a large OS 
distribution the size of Fedora, and previously Red Hat Linux.  I 
don't remember a release ever that wasn't delayed a slight amount 
be it a day or a few days, etc.  The only difference here, is 
that in the past Red Hat did not officially announce the public 
release day of the operating system, so the official release was 
"when it is ready", and so if there was a need to ever delay a 
release by a day or two, or three or four or whatever in the 
past, you never were any the wiser now were you?  ;o)

In the spirit of openness, and as a showing of commitment to 
being open, you have now been priveledged to have previously 
been given a hypothetical target-release-date by Red Hat, which 
was in past OS releases not given out to anyone ever, and for 
this very reason - that people go hyper paranoid if a date is not 
met exactly.  All manner of FUD gets spread now doesn't it.  ;o)

And so, now people can perhaps understand /one/ of the perfectly
rational reasons why Red Hat did not previously preannounce OS 
distribution release dates, etc.

So what has happened then?  Oh my gosh!  Some last minute
surprises were found that demanded Red Hat attention to fix prior
to being released!  Oh wait a minute...  That's not news, thats
more or less to be expected in any major OS project's release
planning.  In fact, I'll go as far as to say, every release ever 
made from now on, will probably miss it's publicized target date 
for one reason or another.  Why?  Because that is the way it is 
with software development.  No matter what date you set, you 
always have a couple of unexpected surprises pop up.  Not 
necessarily major ones, but big enough to warrant fixing in order 
to have a nicer release.

I think the majority of users appreciate this openness of the 
Fedora Core development model very much, and are very happy to 
see that Red Hat considers Fedora Core an important enough 
release to delay it for last minute problems like this, and from 
the majority of people's thoughts that I've seen today, I think 
most people are happy to see this, even if they're a bit 
disappointed they have to wait a couple more days.

What is a couple of days though in the big picture really?  I 
mean, we could fake the dates and add in a week of buffer time, 
so that everyone thinks the release is on the 10th of November, 
but then plan everything and work on everything to hit a date a 
week earlier, but then that's not exactly being open about the 
project's development now is it?  ;o)

With this public knowledge being provided, comes the 
responsibility by people reading and interpreting the knowledge 
to realize these things are just natural parts of the software 
development process in general. Perhaps parts they were shielded 
from in the past or were not any the wiser about (and didn't need 
to be).  However, with the Fedora Project taking on a community 
link now, and with the goal of having more community 
participation and contribution/interaction in the future, it's 
can't really work unless we let people know these kind of things 
as soon as possible.



>Relating to some previous posts on RedHat moving to
>enterprise...
>
>How many of you that are currently running RH up to version 9 on
>many servers are now planning to move away from RedHat?

Who could really say?  ;o)  From the majority of feedback I have 
seen, most users are more than happy to either switch to one of 
the RHEL product options available, or to continue using the 
freely available OS previously known as "Red Hat Linux" which is 
now known as "Fedora Core" and is being made a more community 
oriented project.

>I would certainly prefer not to, but fedora does not seem like a
>solid bet.  I pay for RH subscritions and have 9 installed on
>about 60-70 servers, and have purchased the box sets before, but
>I dont like the way Fedora popped up.  Are others going to keep
>Fedora on thier servers?

I suspect that depends a lot on what types of service and support
individuals and companies out there require for their machines.  
It also depends on how much money or time they'd spend trying to
support their OS installations themselves.  If people honestly
look at their total cost of ownership for running Linux systems,
and that includes their own personal time, and the time of their
employees, etc.  they should compare these various factors with
the cost of using Red Hat Enterprise Linux very carefully.  Some
users misjudge the value RHEL provides initially until they think
of the real value they get from using Linux in their business,
etc. in the first place, and they think about the true amount of
work and time that Red Hat's Enterprise Products will be saving
them.

I'm sure there are also various users to which RHEL may not meet 
their given needs, and for them, they may wish to look at Fedora 
Core, after fully understanding the purpose and project goals of 
the Fedora Project, and clarifying anything they don't quite 
understand.  Of course there will likely be other options 
available such as 3rd party support, and other distributions too.

Each person should try to consider the various options they have 
available, and choose the one that best suits their needs.  
Hopefully that will be RHEL or Fedora, but if not, hopefully it's 
got Linux inside anyway.  ;o)


>The move to fedora was a mistake for RedHat.

That remains to be seen yet really.  Some people will of course 
be of that opinion, but it's just that really - an opinion.  Many 
people differ greatly in opinion, and believe the Fedora 
Project's goals, and Red Hat's movement in this direction are 
innovative and fairly unique in many ways.  That's individualism 
at work.  ;o)  Feel free to speculate, but only time will truely 
tell.

>I understand why they did it, but they went too far and
>discarded loyal users ( some paying some not.  Perhaps still
>including a box set would have been a good idea.  Although
>looking through the mailing list, is Fedora up to the task of
>running on a server anymore?  I am not sure.

I've come to the conclusion over many years that if you are very 
very big, and you make changes to pretty much anything, you will 
upset some people greatly, and you will please other people 
greatly as well.  No matter what the change is, many people are 
resistant to change and will be against the change without even 
waiting to see if it works out or not.  This is of course simple 
human nature, and is to be expected.  The bigger you are, the 
more you get to see it with every small thing you do.  It's in 
many ways just like politics.  No matter who you vote for, they 
never seem to completely do what you had hoped.  They also can 
never please everyone out there, for every decision they make 
will be loved by some, and hated by others.

In the end, you realize the age old adage is true:  You can 
please some of the people all of the time, and all of the people 
some of the time, but you can't please all of the people all of 
the time.  That cliche holds true today as it ever did.  Once you 
experience the deep truth to this, you ask yourself the next 
question:  "What do I do then if I can't please everyone all of 
the time?"

The answer to that of course, is that you try to come up with 
fresh interesting ideas, try to be innovative, and try to stay on 
top of things, pushing technology here and there, taking a few 
chances here and there, etc.  You come up with some goals, and 
then you aim to achive them and strive to do your best.  You know 
all along you wont produce something 100% perfect, but you know 
you can do a great job which pleases the majority of people in a 
big way, and so you set out to do that.  Everyone will have their 
own quirks and gotchas when everything is said and done, but the 
main thing is that progress and evolution can't occur if you 
stick your head in the sand and plod along on the same old path 
always.  Sometimes you need make major changes and try new ideas 
out.

I feel that the newfound openness and joining with the community 
in such a larger and more direct way, has been needed for a while 
now, and that Fedora now provides the foundation upon which 
everything can build into new grounds of innovation more quickly 
and with a lot of good coming out of the end result.

We can't predict the future, but we can guide the path to the 
future in many ways, and have influence on it's outcome.  With 
the Fedora Project being community based, everyone else can 
participate in the processes now and more and more as time goes 
on and infrastructure is put into place.

Hope this helps to think of things in a more wider and open 
minded sense that perhaps some people hadn't considered before.

Take care,
TTYL


-- 
Mike A. Harris     ftp://people.redhat.com/mharris
OS Systems Engineer - XFree86 maintainer - Red Hat