[Compcomm] My ideas (for the record)

[Mike Dransfield]

Alyssa Hung wrote:
> Mike Dransfield wrote:
>   
>> Alyssa Hung wrote:
>>     
>>> Danny Baumann wrote:
>>>   
>>>       
>>>> Hi,
>>>>
>>>>     
>>>>         
>>>>>> ACK in principle, but there's always the risk that we "discuss things to
>>>>>> death" (such as the project name...) and in the end nothing will be
>>>>>> done. I'm not saying that this will happen everytime something is
>>>>>> discussed, but we should not forget that risk.
>>>>>>   
>>>>>>         
>>>>>>             
>>>>> We need to at least know what the issues are, coming up
>>>>> with nothing IMHO is better than coming up with 2 competing
>>>>> solutions that each dont really fill everyones needs.
>>>>>       
>>>>>           
>>>> Ok, that's something we disagree in then. I think it's better to come up
>>>> with a solution that fills the needs of the vast majority than with no
>>>> solution.
>>>>     
>>>>         
>>> Indeed. Many, many users are grateful for having access to tools that 
>>> make the mystique of configuration comprehensible to them. Many, many 
>>> users also appreciate having access to plugins that are CLOSE ENOUGH to 
>>> do what they wish, even if there are technical limitations that can only 
>>> be properly addressed at a much later time.
>>>   
>>>       
>> This is my problem, it is not necessary to design a different
>> configuration system just to have a user friendly GUI.
>>
>> Users do not understand this distinction so when they hear
>> about people saying libbs is crap then they hear 'gui tool is
>> crap', thats not the case at all.
>>
>> I am fairly sure we have been saying this for a long time.
>>     
>
> I admit to not having researched grahpical Compiz configuration tools 
> extensively, but at the time I started with compiz-quinnstorm, the most 
> graphical tool available was gconf-editor. And that was confusingly 
> divided into settings for "all screens" versus "screen0". There was no 
> plugin dependency resolution.
>   

You can still use this as a fallback in compiz, but it is not
the only means.

I think it is very important to have a fallback because there
have been a few notible cases where people are left without
without features because of a bug in the settings manager.

Editing libbs config files is not user-friendly.

> I don't doubt that better solutions now exist for Compiz. If I recall 
> correctly, the latest generation of beryl-settings was partly inspired 
> by screenshots (or mock-ups?) of a Compiz configuration tool.
>   

Yeah - I forgot that :)

> But the fact is that Beryl got it done FIRST. Perhaps it's not a fair 
> way of passing judgment, but you know that users and developers don't 
> necessarily share the same perspective.
>   

Technically it was done first in compiz-quinn, but if
the fork had not happened it would have been done
in the same time for Compiz.

> Beryl also got an assortment of much-requested plugins done first, such 
> as inputzoom, which cleverly circumvented the need for proper input 
> transformation.
>   
For a moment then, I thought you were going to mention
plugins like animation and group.  You need to work on
identifying your strengths better ;)

Where is inputzoom now?

> "Usable" and "first" make bigger waves amongst the users than "process" 
> and "maintainable". It's perfectly fine for better solutions to 
> supersede obsolete solutions. As long as there's an immediate solution, 
> it's easy to be patient.
>   

Yes - This is perfectly valid, it all really depends on a
case-by-case basis as to what is 'worth it'.

>   
>> This does not make sense at all.  The backend was already
>> ready.  This is a case where it would actually be faster AND
>> produce a better result to use dbus for communication.
>>
>> Have a look at compiz-settings.  IMHO it was way better than
>> anything else I have seen so far.  It took a fraction of the time
>> to produce libbs/ccsp/ppsc-settings etc etc.
>>     
>
> I am no authority on these subjects, but as a user, I have never managed 
> to get DBUS to work. Beyond prepending dbus-launch to the compiz command 
> line, dbus-send and the rest never worked for me. Given enough effort, I 
> can definitely solve the problem.
>   

I expect you are using an old distro, the gnome/kde
session needs to be started with dbus-launch.

The alternative would be to access the backend directly,
even that (which is used by gnome-compiz-settings) is a
better solution than libbs.

> But because libbs was both available and simple (just one more thing to 
> compile; no system-wide configuration, no chance of inadvertently 
> breaking other packages), and because Compiz was the only thing that 
> required DBUS on my system, I haven't been suitably motivated to get to 
> the bottom of my DBUS woes. This will change, too; KDE4 is coming and 
> it's all about DBUS.
>
> But you see what I'm saying about immediate solutions versus proper 
> solutions? Users are not going to jump through hoops to accommodate one 
> program, especially if there's another one that does the same thing, but 
> more easily.
>   

Nobody ever said you can never work on short-term hacks.
Just dont be upset if they have lots of bugs and other problems
and they are thrown away eventually (possibly before the real
solution arrives)

Other developers will point this out but everyone is free to do
what they like.

>> The side effect of using the proper methods is that everyone
>> will be able to use it, not just the small few who install all the
>> dependencies and change their start script.
>>     
>
> Yes. DBUS will become critical to many systems in the near future, and 
> is surely the way to go moving forward. But the near future is not 
> today. Frustrating, isn't it?
>   
I have not seen anyone with major problems with dbus,
it is used in KDE 3.5 too.
> It's ironic how Beryl accumulated so much mind share precisely because 
> it did things impulsively. Compiz was a critical part of the equation 
> even then; if the Compiz developers had quit while Beryl was ahead, 
> where would we be today?
>   
You don't think that possibly some of the 'mindshare' came
because of effects 'inherited' by Compiz?

A lot of heartache was caused because things were not
thought through properly (that includes beryl AND this
merge)

>  From a user's perspective, it may be best for immediate solutions and 
> proper solutions to be developed concurrently.
>   
Yes, nobody has a problem with that.  Its often the
case that the proper solution cannot be done by the
developer doing the quick solution.

Look at my scripting plugin, its the definition of hacky
but I made that clear to people.
>   
>>> Not cult. Culture. ;) Some of us users feel threatened by the notion 
>>> that we'll have to change, maybe change a lot (our posting habits, 
>>> expectations of the sorts of responses we will receive, etiquette, ways 
>>> of eliciting meaningful responses, verboten subjects, dress codes, 
>>> etc.), to integrate well into the Compiz society. Please show me that 
>>> these fears are unjustified.
>>>   
>>>       
>> Who said you had to change any of those?
>>
>> The whole point of these discussions is so that we can
>> find a way to accomodate everyone in one place working
>> towards common goals (or something like that).
>>
>> The only rule would be post in the right category, but I
>> think you have that rule on all forums.
>>     
>
> A cursory glance at forum.compiz.org shows an abundance of posts that 
> are well-researched, concise, and evidential of the poster's 
> professionalism. This is, how do I say... Intimidating? When the 
> majority is so sophisticated, it makes ordinary people feel unfit.
>   
I am sorry to hear that, I am not sure what to say other
than delflick posts there, and I dont *think* he feels
intimidated.
> And though the aesthetics of the visual style have clearly been thought 
> through, it's too clean. Always pristine, as though uninhabited. Do you 
> ever have a feeling, when walking into an extravagant citadel, that you 
> shouldn't touch anything? That's a bit like how forum.compiz.org's theme 
> feels.
>   

Yes maybe you are right, I remember RYX mentioned at
the time that maybe it could do with a bit more colour.

The way to change it would be to post a request for a bit
of a facelift.  If you can help with some mockups then all
the better.

LOL :)))  :!!!1 :D
> Unintentional, perhaps. But implied nonetheless.
>
>
> ~Alyssa
> _______________________________________________
> CompComm mailing list
> CompComm at Rock3d.org
> http://www.ubaight.com/mailman/listinfo/compcomm
>