Will there be a boot splash

[R S Gill]

Totally agree with your sentiments!

The day Linux is made for users, is the day Linux will take over the 
desktop.

Gill

June Tate wrote:

>On Mon, 2004-10-25 at 14:48 -0400, Brian Barr wrote:
>  
>
>>ubuntu-users-request at lists.ubuntu.com wrote:
>>
>>    
>>
>>> Those of us who *can* troubleshoot
>>>are perfectly capable of getting to the information; everyone else is
>>>just intimidated by the endless rows of cryptic text.
>>>      
>>>
>>Intimidated?
>>    
>>
>
>Intimidation was probably the wrong word to use here. More like wary or
>even maybe disgusted. In all honesty, these debug messages are 99% used
>by development teams and kernel hackers -- average users don't need to
>see them, and they only add to the "ugliness" that Linux is generally
>perceived as.
>
>Like it or not, how "pretty" the interface and operating system is is a
>major factor with regards to adoption. Take a look at former Linux users
>who have switched to OS X -- they didn't do it for the simpleness of the
>hardware, they did it because it was *NIX with a clean, well polished UI
>that didn't have the "hodgepodginess" that traditional *NIXes had (eg:
>IRIX and Solaris).
>
>Why do we have to mar the cleanliness that we have in GNOME and GDM
>already with the diagnostics of startup scripts and the kernel? When
>something goes wrong it's usually a pretty obvious thing, so in response
>we have a boot menu option for a "rescue" mode (which, incidentally
>replaces the boot logo with the diagnostics again). Also, since it's
>toggled with a boot flag on the kernel's command line, those of us who
>actually _want_ to see those diagnostics can simply edit
>the /boot/grub/menu.lst file and remove the flag that turns it on in the
>first place.
>
>Besides, when I go to demonstrate this operating system to a Windows
>user or a Linux newbie, the first thing they will usually see is all of
>these ugly diagnostic messages. My Mom (whom I have managed to get to
>try Ubuntu) recently commented to me, "Oh, so you're getting those error
>messages, too. They must not mean anything, then," when she saw my
>desktop boot up Ubuntu.
>
>Users see the word "error" or "warning" anywhere, and they immediately
>become worried that something is wrong needlessly. A perfect example is
>bug 1869 ( https://bugzilla.ubuntu.com/show_bug.cgi?id=1869 ) -- shpchp
>and pciehp both are being modprobed into the kernel to test if hot
>pluggable PCI slots are available. Since they don't exist, they dump a
>bunch of error messages -- error messages that more than one person new
>to Ubuntu has reported as "true errors" because they didn't know any
>better.
>
>  
>
>>How about adding a nice text message as it starts up saying:  
>>"And now we will start the boot process...you can probably ignore most of this."
>>    
>>
>
>But with the amount of diagnostic information that is dumped out to the
>screen, this line would most often scroll by without most people
>noticing, and it still doesn't solve the "hodgepodginess" issue I
>outlined above. Even if users _do_ see the message, it will only add
>confusion into the mix, since the generally accepted idea is that if an
>error is reported then something is wrong -- we'd be contradicting
>ourselves (as an open source software community, that is) and sending
>mixed signals to the users.
>
>  
>
>>Once you see that start up phase once or twice, it becomes "normal".
>>    
>>
>
>Is it? When the people who reported bug 1869 saw their error messages
>over and over, was it normal to them? Obviously not because it's not
>related to any other errors in their systems, and again, "...you can
>probably ignore most of this" isn't going to stick around on the screen
>for very long. Besides, using the message above doesn't reduce the
>concern for the user: these error messages were obviously written for
>_some_ reason and aren't being dumped needlessly (which is true, for the
>most part).
>
>  
>
>>I would think part of this "humanity" thing is to not treat your customers like
>>fools.  Raise your expectations, and people generally meet them.
>>    
>>
>
>Agreed, but you have to keep in mind that there are three major factors
>here:
>
>- By keeping the diagnostics, we reduce the professional polish and 
>  cleanliness that the distribution has -- going from diagnostics 
>  scrolling by to a nice clean GDM screen is a major break between 
>  styles.
>- By keeping the diagnostics and placing the "...you can probably 
>  ignore most of this..." message on the screen, we're not reducing 
>  user's concerns by any amount. In fact, we're only adding confusion.
>- By displaying error and warning messages on screen, we're generating 
>  excessive bug reports and list traffic.
>
>This really has very little to do with treating users like fools. It has
>everything to do with treating our users like users and not developers.
>Most of these diagnostic messages users don't need to see because it's
>debug output for the developers (like the afore mentioned bug). It's the
>same reasoning as to why compilers are not included in the default
>software selection: _users_ don't need that kind of software --
>_developers_ do.
>
>--
>June Tate * june at theonelab.com * http://www.theonelab.com
>
>  
>