Page 5 of 50 [747 Posts]   Goto page: Previous 1, 2, 3, 4, 5, 6, 7, ..., 48, 49, 50 Next
Author Message
the last saviour
PostPosted: Wed 18 Sep 2013, 07:48    Post subject:

Lieber Herr RSH;
I'd just noticed a funny calendar in Pwidget. As you can see the last coloum should be one row up.
I know that LaZy project is finished, but in case you will still continue Pwidget in other release.

This is also an argument to change my focus. I don't want to be responsible for problems and issues that are results of remastering a puppy and/or adding/removing stuff to the OS etc.pp. My responsibility in the future should only be related to the programs developed and published.
---- This is the indication that you need a team work!

Believe me: not a single second was wasted!
---- I believe that. But your time value will be markedly increased if you product is appreciated by a bigger scale of users.

Before I continue I have to correct a mistake, very sorry for this;
From If it's for their personal use that's should be fine.
To If it's for a group of users that's should be fine.

Did this come from an office memo in the Mc Carthy Witchhunt Trials of the 50'
---- Sorry, I've never heard of that story.

the last saviour; Expressing one`s self. And helping others.
---- As God has assigned me to be the Last Saviour, helping others is my duty definitely. Smile
I don't express myself, but I did interpret my 5th dream. God had already created 5 colourful dreams for me. 4 dreams had already become true. And I will not surprised if the last one will be true, it's more colourful than others.
Dreaming is the most important method for divines to give information to somebody since the ancient time. Normal dream is the outcome of data hidden in the subconscious. But because of the limitation of memory and they're going to be destroyed(forgot), all these data are incompleted. So the normal dream is B&W and fragmented, while the continuous colorful dream is going to be true, because some divine creates that dream for you. This is the major method that God gave information to someone in India. Thus many truths are hidden in Hindu scripture. But there are also many mistakes because of the wrong interpretations.
I found this website 3 months ago, firstly it couldn't convince me. But now I believe in some information.
If Rockyfeller's cars are pink and sky blue. Then I believe that the building I saw in my dream is the Rockyfeller building. And they are not ET but chipped.

What makes Debian and Ubuntu work? Lots of folks each doing a small part.
They aren`t one-man-shows, like it is here at Puppy.
But as many say... This is probably all Puppy will ever be.
---- I believe Mint, Fedora, Mandriva and Android also has a team work. Not to mentioned that MS, Apple, Samsung, Nokia etc. hires dozen of programmers. One-man-Puppy is only for a group of users not for the top commercial market level. But team-worked Puppy can achieve the top commercial market level.

Because they vainly hope it will bring in new ideas but eventaully they realise that having it for your own use is as far as you will get.
---- Yes, it will bring the new ideas. But the new ideas can be etiher better or waste of time. I had tried many variants of DM&FM in Linux and found KDE+Dolphin to be better than MS, and XFCE+??? is slightly better than MS.
Should we spend more time to create any new idea of DM&FM to be better than KDE+Dolphin?

Puppy is at the top of it's class for a minimal sized distro. The creativity and innovation shown by the community has been nothing short of phenomenal.
---- But as CPU speed can be double every ~5 years and media storage can be double at a faster speed, there's no significant point to pay attention to OS size and speed. MS has used the same DM&FM for >20 years. Is there any user complain of this old DM&FM? MS only adds more features and fixed those bugs. And users enjoy MS because of the vast variety of softwares and hardware friendly. As you probably know that many Puppy users still use MS because of features of software and hardware support is better.

At last I have to inform you that there's a big knocking of unknown source in my room while I explain about alien in this comment.

Best regards.
PostPosted: Thu 12 Sep 2013, 11:08    Post subject:

Hmm I played with a distro with a 3.2.xx kernel and similar dated binaries...was a nightmare... backward compatability is not a Linux strong point and more bugs than a compost convention. In puppies case the wrappers seem to bog it down increasingly with each release.

My website host has recently updated to a 2.6.33 kernel.... servers are pretty conservative when it comes to what they run as they need it to work 24/7...and some devs new core system experiment would ruin their day. Tried and tested is the sensible approach in the commercial world.

actually greengeek the daily system(s)are circa puppy 4.12/slax 6/lenny .... still ok for recent software (amazing how the need for dependancy updates disappear when you compile applications yourself) and building drivers for such has generally been no problem. The 2 series got to be a little awkward in some areas...though it was a testing ground for all those things that ended up on the newer systems. I do have a rejiggled Lucid for some point in the future.

All things will change.... just at a more natural pace I prefer. I don't consider 6 months as 'old' or 'ancient' ...what would that make me !!!!... Shocked At least there is a pile of willing beta testers out there preparing my future....

A toast to Lazy Puppy ... I always liked the name Smile

PostPosted: Thu 12 Sep 2013, 05:02    Post subject:

sunburnt wrote:
Older versions don`t tend to be better
If thats true then why is mikeb still running on a v2 puppy? Must be a reason why he hasn't gone to a 5 series puppy.

And look at what RSH is doing with puppy - in a way he is "de-inflating" it. By using the sfs approach he is enabling a tight core, with the specific additions the user wants.

I'm referring to the idea of having a technique of sticking with the smallest, fastest kernel that DOES support your hardware, and adding on only the required sfs files for the newer software functions you want.

As hardware marches on, maybe it is not the entirety of puppy that needs to keep pace - maybe only the kernel and sfs loader or ROX-app loader that need to be updated. So you only need to update if your hardware fails to work with the previous kernel/loader combo.

Still just spitballin' here...
PostPosted: Thu 12 Sep 2013, 02:35    Post subject:

Older versions don`t tend to be better. New ones support newer stuff and the old also.
The older O.S.s and software get the less compatible and more worthless they are.

Like Apple, if you have fixed known hardware, then compile the driver modules into the kernel.
This is the way Linus Torvalds intended it to be. Supporting all types of hardware is a nightmare.
PostPosted: Wed 11 Sep 2013, 18:37    Post subject:

Hmm you describe something between an apple mac and those linux netbooks that appeared for a while. The latter had such as suse on them using older kernels and drivers built specifically.

I am one for compiling drivers on established kernels and as you mention trying to keep up with hardware changes is producing some humungus kernels/modules/firmware builds. My feelings on the subject would be to modularise at least wifi drivers ie more akin to the windows approach of installing drivers to suit... even with video this sort of happens if you have Nvidia or Ati. This thought does tend to fall in a heap when you are dealing with multiple kernel changes every week though. I wonder if such as Lucid have had such popularity partly through sticking with one kernel for a sustained period. As someone pointed out a kernel change is akin to a new release of windows.

Are we off topic ... usually and apologies to RSH et al but I suppose the pros and cons of 'distro fever' are related.

wear loose clothing

PostPosted: Wed 11 Sep 2013, 18:14    Post subject:

Thanks for the answer mike. I certainly see the value in keeping one's custom pup under wraps to avoid it soiling the floors of people who run completely different hardware...
I do sometimes wonder if a good approach to getting a well-functioning pup would be to have a project built around a specific hardware. For example your custom pup obviously works well for the hardware you use it on (including good boot speed I imagine..). It seems to me that part of the reason for puppy's fragmentation is that the hardware keeps changing so the development of any given puppy never quite stabilises.

I'd like to see a project to make an old puppy perform like magic on something like one specific model of an older HP laptop (I can't say specifically which one would be good for the process..).

Imagine if a 2 series pup could be made to run real quick on such a machine - surely it would only need a basic, speedy core with a modern browser sfs and office sfs added on? Does it really need a 5 series pup weighted down with modules and modifications that allow it to run on a wide range of hardware??

Just spitballin' here.
PostPosted: Wed 11 Sep 2013, 16:13    Post subject:

@greengeek ... Erm No and sort of.... Intially changes were made to optimise and try out alternative saving methods and sfs loading. I tended to keep things 'the puppy way' but eventually it becomes necessary to break away from the convention. Saying that when it comes to the initrd I built around the 2 series as a simpler base after finding 4+ increasingly hairy so to anyone familiar with older pups its not that no sub sub sub sub folders and numerous kernel parameters to try and circumvent a slow boot process..but those only used to puppy 5 would find things uncomfortably different..

Of course it also becomes the case that anyone doing this will only have the hardware at hand to test on eg in Lucid usbhid drivers were split up and involved script hacking to deal with and if its not right wireless input devices might go awry for example but I have none to verify all is still happy. I have no dial up modems so cannot check those, though I tried to leave such alone, if a kernel has been swapped they may be affected (modprobe backend and related scripts)
I add certain proprietry drivers as needed. I avoid bleeding edge.

In other words such projects tend to become custom jobs though I do have a small menagerie here that would beat me up if the machines did not behave Very Happy It also means my scripts become less and less compatible as puppy marches on.

Anyone releasing their baby on the puppy public is exposing themselves and it can be a thankless task... I have only had a small taste of that so generally I tend to throw in and discuss ideas so perhaps others might want to give them a whirl on their projects.

A long winded vague answer for you....

PostPosted: Wed 11 Sep 2013, 16:01    Post subject:

To use LazyPup as an example, just a few are needed to maintain a bare-bones core O.S.
And folks each specializing in a chosen group of new app. packages, and so doing it well.
Web browsers, media players-editors-encoders, office suites, games, languages & dev., etc.
Once there`s a good set of app. groups, then the only work is building newly released apps.

With about 6 people ( and 4 substitutes so no one`s over worked ) this could be sustainable.
This is how Debian and Ubuntu maintain their distros. And they`re very large cooperatives.
This type of organization could work even better in smaller groups like is needed for Puppy.
It depends so heavily on the people involved, but limiting the work load for each sure helps.
PostPosted: Wed 11 Sep 2013, 15:33    Post subject:

mikeb wrote:
I have taken all the Puppy initrd/boot/wrapper/shutdown scripts apart and rewritten or tidied as appropriate adding features in the process.
Hi Mike, have those changes found their way into any mainstream pups or were they changes made for your own use? I'd be keen to try an iso with such updates.
PostPosted: Wed 11 Sep 2013, 13:47    Post subject:

In Puppy`s case, it wouldn`t take many people to form a group.

how many exactly... I heard three's a crowd Very Happy

I got the Lucid initrd init down from 60K to 25k with a similar range of options so theres always room for optimising.

The main releases are definately the work of several individuals though one man shows can sometimes produce good results... half the problem is testing with the huge range of hardware and software out there and thats where input is needed. The murga forum at least does provide a good source of feedback for anyone attempting developement.

Perhaps what I see lacking is an effective way to contribute ideas to the main releases which as I suggested earlier is why puplets often appear, not including the ones which are just standard releases with 2 added programs and a nice wallpaper.
How much of the ideas in Lazy puppy will actually ever get utilised in precise??. And bear in mind its dev has been inspired by other older puplets showing the possibilities...

PostPosted: Wed 11 Sep 2013, 13:32    Post subject:

Puppy`s boot process is complex because of it`s many boot types.
And the union file system adds more complexity to all aspects.

# In line with RHS`s retirement:
Puppy is for the most part a one-man-show. Many of them...
This doesn`t lend itself to preserving the improvements made.
And greatly limits what can be accomplished. The efforts of one.

# Doesn`t it seem that cooperation for common good is best?

Debian and Ubuntu are cohesive as the individual`s load is small.
So the participants remain doing what they do best for the whole.

# In Puppy`s case, it wouldn`t take many people to form a group.
PostPosted: Wed 11 Sep 2013, 10:42    Post subject:

Well since puppy is held together with a plethora of scripts their condition is relevant to the description of Puppy as 'simple'...not just my opinion ask anyone who has worked on it.

The binaries are taken from other distros with few exceptions so any 'simplicity' in that domain has nothing to do with Puppy.

I have taken all the Puppy initrd/boot/wrapper/shutdown scripts apart and rewritten or tidied as appropriate adding features in the process. I may not be a 'programmer' but i do have a chunk of experience in this particular area.

PostPosted: Wed 11 Sep 2013, 09:54    Post subject:

mikeb wrote:
The beauty is in the simplicity.

Ah so you have never examined puppies scripts.
For simplicity look at slitaz or slax.....

Scripts are non-compiled code. If you think you can improve a script, I'd suggest you learn how to do that vs writing meaningless comments about everyone elses' work.
PostPosted: Wed 11 Sep 2013, 06:53    Post subject:

The beauty is in the simplicity.

Ah so you have never examined puppies scripts.
For simplicity look at slitaz or slax.....

As for pup most changes I see in the main releases are the ones implemented by Ubuntu. The real innovations are seen in the puplets and community releases.


@sunburnt...the neat stuff is used here on a daily basis Very Happy
PostPosted: Wed 11 Sep 2013, 01:48    Post subject:

sunburnt wrote:

But as many say... This is probably all Puppy will ever be.
Thomas Alva Edison:
No new idea will ever prosper in an existing company.
A new company must be formed around it.

Puppy is at the top of it's class for a minimal sized distro. The creativity and innovation shown by the community has been nothing short of phenomenal.

Puppy isn't a company, so there's not much in the direction of innovative breakthroughs with broad appeal. It's more like a language, that can easily be adapted for individual use. The beauty is in the simplicity.
Display posts from previous:   Sort by:   
Page 5 of 50 [747 Posts]   Goto page: Previous 1, 2, 3, 4, 5, 6, 7, ..., 48, 49, 50 Next

Powered by phpBB © 2001, 2005 phpBB Group