3:32 p.m.
Hi
http://www.nascencetech.com/newvillageboy/2005/09/05/full-review-fedora-c...
If anyone is willing to pick this up and respond to the comments, go
ahead. Not much to tackle here
regards
Rahul
9:22 p.m.
On 9/12/05, Rahul Sundaram <sundaram(a)redhat.com> wrote:
Hi
http://www.nascencetech.com/newvillageboy/2005/09/05/full-review-fedora-c...
If anyone is willing to pick this up and respond to the comments, go
ahead. Not much to tackle here
here's my email to the author
<begin transmission>
I just read your review and I'd like to take a moment to give you some
background and updates on some of the negative points you bring up
about your fc4 experience. Its a pretty well reasoned review overall,
and I appreciate the effort you made to identify your personal
experience with KDE and potential personal bias right upfront.
"This being my first experience with Fedora Core, I was surprised and
annoyed to find that the only native Linux filesystems available for
selection were ext2 and ext3. My main Linux machines are running
Gentoo and Kubuntu and all of them are either using ReiserFS or XFS.
The main reason for giving up ext3 is performance, or rather the lack
of it. A more detailed analysis of this can be found here."
Making a choice as to which filesystems to support is a complicated
decision. Performance in different circumstances is one thing to
consider, but its also important to look at maintainability as well as
integration of the filesystem with other aspects of the system. How
well is the filesystem supported in the mainline kernel? How well does
the filesystem support other key aspects of the operating system? I'm
not going to try to debate benchmarks with you, instead I'll point out
that the default filesystem in Fedora needs to be able to work
reliably with other features such as selinux. The alternative
filesystems you mention each have their own strengths, but fail to
meet the integration and reliability needs of Fedora Core as a whole.
As you can see from the unofficial fedora faq, people can get access
to these filesystems but they aren't presented to users by default. I
hope you appreciate the logic in that. For the unsuspecting user who
doesn't know anything about filesystem differences, it makes the most
sense to streamline the install process with the filesystem that has
the confidence of the distribution developers who will be responding
to problems. For more experienced users who can support themselves or
who can work with upstream if a problem arises, we can definitely do a
better job of documenting how you get access to those additional
choices. The process for how the release notes are written has
undergone a transition and will be under more direct control from
active community writers. Some of the fedorafaq.org material about how
to enable this alternative filesystems could very well make it into
the release notes for fc5 in the new process:
http://www.fedoraproject.org/wiki/DocsProject/ReleaseNotes/Beats
You'll notice there is a "filesystems" beat listed and a slot open.
You wouldn't happen to know someone interesting in patrolling that
beat would you?
"As mentioned above, FC4 lacks certain applications and libraries that
do not come under any bona fide Open Source Initiative-approved
licenses. A number of these are pervasive enough that they have become
essential parts of any desktop system:
1. Java support
2. Macromedia Flash support in Firefox
3. MP3 support
4. Producer 3D support for NVidia and ATI cards"
This is a much tougher issue because of Red Hat's managerial role, the
project has to respect the opinion of Red Hat's legal counsel and
avoid opening up legal liability for Red Hat. I assure you, no one is
particularly thrilled with not being able to direct users to 3rd
party's who provide the applications. Okay well, I'm sure there are
certain developers who would love it if addon hardware drivers would
stop being used completely because the amount of system instability
and bugreports that result...but we all know thats not going to stop
happening. Yes, Fedora Core and Fedora Extras are dedicated to
providing 100% open source functionality but the project also realizes
that there are gaps in the technology we are able to provide. The
project is trying to walk the line as best it can, and there have been
on-going conversations to better define exactly what that line is. One
of the results of discussion is the FC4 release note errata which
provides a link to fedorafaq.org for users to find answers to commonly
asked questions.
http://fedora.redhat.com/docs/release-notes/fc4/errata/
I expect a similar pointer to an unofficial faq to be provided in the
fc5 notes unless of course the legal landscape changes in the
meantime.
"One MAJOR grouse I have is the inclusion of the up2date icon that
appears automatically after installation. It maintains that I have no
updates necessary, which I found suspicious as FC4 had been released
for almost 3 months. Further checking on the Unofficial Fedora FAQ
website revealed that up2date still works but "it no longer uses the
Red Hat Network (RHN). However, with Fedora, it's better to use yum to
get updates, instead of up2date." Now, that sounds like a good reason
as any to dump up2date and just create a GUI front-end for the command
line tool yum instead. Issuing "yum update" for the first time
revealed a whopping 131 packages to be updated, including some rather
critical ones for Firefox, for example. In this respect, the up2date
status notification that "no updates (are) available" sounds almost
malicious."
Sadly, there is a bug in up2date/rhn-applet that is misparsing the
configuration files. The ability to use the same metadata and
configuration files as yum was added before fc4 but there was an
unexpected problem found after fc4 branched from development.
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=161071
I have not personally seen a written communication saying that
up2date/rhn-applet are going to be removed in fc5, but I do believe
people are working on a replacement gui. If I had a better sense as to
where development on the new codebase stood I'd give you a reference
to it.
There has also been an effort to better document how to use yum in the
near term.
http://fedora.redhat.com/docs/yum/index.html
I hope these answers have helped. With your permission I'd like to
repost this conversation to the fedora-marketing list to make sure
that the marketing team has a chance to correct any factual mistakes
and to make sure I have not mis-interpreted the project's position on
any of these issues.
-jef
7:39 p.m.
On Mon, 2005-09-12 at 22:22 -0400, Jeff Spaleta wrote:
On 9/12/05, Rahul Sundaram <sundaram(a)redhat.com> wrote:
> Hi
>
> http://www.nascencetech.com/newvillageboy/2005/09/05/full-review-fedora-c...
>
> If anyone is willing to pick this up and respond to the comments, go
> ahead. Not much to tackle here
here's my email to the author
+1 to Jeff for your excellent knowledge of available documentation and
for recruiting for relnotes beat writers. That's the stuff!
- Karsten
--
Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/
gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41
Red Hat SELinux Guide
http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/
7:44 p.m.
Karsten Wade wrote:
On Mon, 2005-09-12 at 22:22 -0400, Jeff Spaleta wrote:
>On 9/12/05, Rahul Sundaram <sundaram(a)redhat.com> wrote:
>
>>Hi
>>
>>http://www.nascencetech.com/newvillageboy/2005/09/05/full-review-fedora-core-4/
>>
>>If anyone is willing to pick this up and respond to the comments, go
>>ahead. Not much to tackle here
>
>here's my email to the author
+1 to Jeff for your excellent knowledge of available documentation and
for recruiting for relnotes beat writers. That's the stuff!
+1 from me as well very well written.
Marc
7:59 p.m.
On 9/14/05, Karsten Wade <kwade(a)redhat.com> wrote:
+1 to Jeff for your excellent knowledge of available documentation
and
for recruiting for relnotes beat writers. That's the stuff!
Other than my..shall we say..defensive..use of the call for beat
writers as a subtle variant of "put up or shut up" is anyone else
actually pounding the streets for people.. pressing the flesh? bashing
heads? blackmailing? making grandiose and misleading promises?
The beat writer stuff seems to be pretty reasonable sized chunks of
responsibility. Where exactly do we hang the virtual "help wanted"
sign to get as many eyeballs as possible looking over that last as
possible?
How much does a Google text ad go for? Cuz man.. how many fedora
meatheads are like me.. and are using gmail for all this list
traffic... can we actually find a way to get project help wanted
adverts into their gmail interface?
-jef
12:33 a.m.
On Wed, 2005-09-14 at 20:59 -0400, Jeff Spaleta wrote:
On 9/14/05, Karsten Wade <kwade(a)redhat.com> wrote:
> +1 to Jeff for your excellent knowledge of available documentation and
> for recruiting for relnotes beat writers. That's the stuff!
Other than my..shall we say..defensive..use of the call for beat
writers as a subtle variant of "put up or shut up" is anyone else
actually pounding the streets for people.. pressing the flesh? bashing
heads? blackmailing? making grandiose and misleading promises?
I'm working on blackmail and head-bashing from within Red Hat.
Also, working on a few infrastructure details before making YAGeneral
call for release notes and beat writers.
Sub-project leaders need to be accountable and responsible for getting
the content into the designated locations (see below). They can
delegate the responsibility to someone in their project to be the beat
writer.
The FDP is taking this content, editing it, converting it into single-
source XML, working with translation, packaging it into fedora-release,
and posting it on the Web.
The beat writer stuff seems to be pretty reasonable sized chunks of
responsibility. Where exactly do we hang the virtual "help wanted"
sign to get as many eyeballs as possible looking over that last as
possible?
I'm just about to make an announcement about this. That we need more
than just some notes dropped into bugzilla, but actual ownership from
the projects.
We're going to open up content submissions to include the Wiki, with a
more open editing rule (have to create an account but don't require
EditGroup access.) Hopefully we'll see more raw chunks of content made
available. We'd see input from testers who might not otherwise mess
with the Wiki.
How much does a Google text ad go for? Cuz man.. how many fedora
meatheads are like me.. and are using gmail for all this list
traffic... can we actually find a way to get project help wanted
adverts into their gmail interface?
Now that's a good idea.
- Karsten
--
Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/
gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41
Red Hat SELinux Guide
http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/
9:04 a.m.
On 9/15/05, Patrick Barnes <nman64(a)n-man.com> wrote:
Looks to be out of reach.
day in and day out... yes... out of reach.
But if we get our ducks in a row and we try to have a "call to arms"
advertisement plan synced up with test1 of a release cycle for even a
single day it might be worth the resource outlay...especially..if we
can turn test release reviewers into our unsuspecting PR minions. So
among all the complaints or fanboy activity in those reviews..they
sneak in a useful blurb about the call to arms for tasks for that
release cycle. So i don't have to personally follow up and
double-dog-dare people to chip in. Can be have a blinding supernova of
"help wanted" advertising on test1 release day?
My first question is can we actually get our ducks in a row in time
for test1 so that if we make a general call to arms for help for
specific identified tasks? Basically making sure the process for each
identified task is ready to grind them up and actually use them right
away. It's not just the release note beat writer task that I'm
thinking about here... though it serves as a good test case example.
There are a number of interesting tasks that could use with a fresh
infusion of blood every release cycle..with the expectation that
you'll encounter a high turnover rate inside the taskteam simply
because accomplishing the task has sucked all the life out of the
unsuspecting participants and left them dry husks which over time will
form a reef-like foundation on top of which a vibrant ecosystem of
teeming Fedora project activity is supported.
-jef
10:25 a.m.
On Thu, 2005-09-15 at 10:04 -0400, Jeff Spaleta wrote:
On 9/15/05, Patrick Barnes <nman64(a)n-man.com> wrote:
> Looks to be out of reach.
day in and day out... yes... out of reach.
But if we get our ducks in a row and we try to have a "call to arms"
advertisement plan synced up with test1 of a release cycle for even a
single day it might be worth the resource outlay...especially..if we
can turn test release reviewers into our unsuspecting PR minions. So
among all the complaints or fanboy activity in those reviews..they
sneak in a useful blurb about the call to arms for tasks for that
release cycle. So i don't have to personally follow up and
double-dog-dare people to chip in. Can be have a blinding supernova of
"help wanted" advertising on test1 release day?
My first question is can we actually get our ducks in a row in time
for test1 so that if we make a general call to arms for help for
specific identified tasks? Basically making sure the process for each
identified task is ready to grind them up and actually use them right
away. It's not just the release note beat writer task that I'm
thinking about here... though it serves as a good test case example.
There are a number of interesting tasks that could use with a fresh
infusion of blood every release cycle..
Fedora automatically gets press when the test releases go out, and will
also get some when the Foundation is announced, so if there is a good
set of pages on participating the URL could be promoted on the press
releases.
Other channels I could think of, which could be used at pretty much any
time:
- fedora-list
- fedora-announce
- FedoraNews
Also possible, but time-sensitive:
- Red Hat Magazine
- print Linux magazines (I'm sure that at least one I've seen has a
"Help Wanted" section for Open Source projects)
- the release notes themselves (there was a help wanted sign on the
FC4t1 notes)
--
Stuart Ellis
stuart(a)elsn.org
Fedora Documentation Project: http://fedora.redhat.com/projects/docs/
GPG key ID: 7098ABEA
GPG key fingerprint: 68B0 E291 FB19 C845 E60E 9569 292E E365 7098 ABEA
11:06 a.m.
On 9/15/05, Stuart Ellis <stuart(a)elsn.org> wrote:
- Red Hat Magazine
- print Linux magazines (I'm sure that at least one I've seen has a
"Help Wanted" section for Open Source projects)
speaking of print magazines... other than the classifieds...how much
does an actual ad cost?
- the release notes themselves (there was a help wanted sign on the
FC4t1 notes)
Until the release notes include slick artwork.. or some sort of
mind-numbing mesmerizing animation..its going to continue to be an
under-read resource for first glance enticement... no matter how
chockfull of useful information it actually is. Its useful
documentation and thus I assume most people simply will not get around
to reading it unless they actually run into a problem and need to
refer to it. But yes of course, the help wanted call goes there to.
Actually i would expect a more verbose description to be in the
release notes which can be refered to from the more hit-and-run
advertising efforts.
-jef
12:33 p.m.
On Thu, 2005-09-15 at 12:06 -0400, Jeff Spaleta wrote:
Until the release notes include slick artwork.. or some sort of
mind-numbing mesmerizing animation..its going to continue to be an
under-read resource for first glance enticement... no matter how
chockfull of useful information it actually is.
We need to decide what goes above the fold for the release notes. Is it
worth hacking on the XSL to get the ToC moved down a bit? We could then
have really important stuff at the top:
* A recruiting item
* Pointer about ForbiddenItems
* what else?
- Karsten
--
Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/
gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41
Red Hat SELinux Guide
http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/
12:41 p.m.
On Thu, 2005-09-15 at 10:33 -0700, Karsten Wade wrote:
On Thu, 2005-09-15 at 12:06 -0400, Jeff Spaleta wrote:
> Until the release notes include slick artwork.. or some sort of
> mind-numbing mesmerizing animation..its going to continue to be an
> under-read resource for first glance enticement... no matter how
> chockfull of useful information it actually is.
We need to decide what goes above the fold for the release notes. Is it
worth hacking on the XSL to get the ToC moved down a bit? We could then
have really important stuff at the top:
* A recruiting item
* Pointer about ForbiddenItems
* what else?
What's New/Experimental/particularly requiring testing ?
--
Stuart Ellis
stuart(a)elsn.org
Fedora Documentation Project: http://fedora.redhat.com/projects/docs/
GPG key ID: 7098ABEA
GPG key fingerprint: 68B0 E291 FB19 C845 E60E 9569 292E E365 7098 ABEA
12:56 p.m.
On 9/15/05, Karsten Wade <kwade(a)redhat.com> wrote:
We need to decide what goes above the fold for the release notes. Is
it
worth hacking on the XSL to get the ToC moved down a bit? We could then
have really important stuff at the top:
Maybe you missed my point. I do not subscribe to the theory that "the
really important" stuff is garunteed to pull people in on a first
glance. If its meant to be a reference document.. then ToC comes
first. If its meant to be a grab your attention document...then
you'll add the appropriate eye-catching cleverness to sucker punch the
audience into reading the ToC as required. Perhaps the National
Inquirer approach.. "Elvis's alien love-child wants YOU to help with
Fedora!"
-jef
5:20 p.m.
On Thu, 2005-09-15 at 13:56 -0400, Jeff Spaleta wrote:
On 9/15/05, Karsten Wade <kwade(a)redhat.com> wrote:
> We need to decide what goes above the fold for the release notes. Is it
> worth hacking on the XSL to get the ToC moved down a bit? We could then
> have really important stuff at the top:
Maybe you missed my point. I do not subscribe to the theory that "the
really important" stuff is garunteed to pull people in on a first
glance. If its meant to be a reference document.. then ToC comes
first. If its meant to be a grab your attention document...then
you'll add the appropriate eye-catching cleverness to sucker punch the
audience into reading the ToC as required. Perhaps the National
Inquirer approach.. "Elvis's alien love-child wants YOU to help with
Fedora!"
At this point, not having the first thing they see be the legal
boilerplate is a major victory.
I understand where you are coming from, and sure, we can't get
everyone's attention without a TV commercial.
But a nice list of bullets that address very common problems or
introduce cool stuff, that's just plainly much better than what we've
had.
Something like this at the top of the page:
http://fedora.redhat.com/docs/release-notes/fc4/errata/#sn-splash
It needs to be reworked to bring a list of items to the top.
- Karsten
--
Karsten Wade, RHCE * Sr. Tech Writer * http://people.redhat.com/kwade/
gpg fingerprint: 2680 DBFD D968 3141 0115 5F1B D992 0E06 AD0E 0C41
Red Hat SELinux Guide
http://www.redhat.com/docs/manuals/enterprise/RHEL-4-Manual/selinux-guide/
6808
days inactive
6812
days old
marketing@lists.fedoraproject.org
13 comments
6 participants
participants (6)
-
Jef Spaleta -
Karsten Wade -
Marc Wiriadisastra -
Patrick W. Barnes -
Rahul Sundaram -
Stuart Ellis