Future Puppy Proof of Concept
The future puppy that I am trying to imagine is just like current puppy, lean and fast, friendly and fun, installing in a variety of ways (live-cd, frugal, etc.) on a wide variety of hardware. The aspect that I imagine developing is that the Repository contains a a larger collection of programs, mainly good linux programs that have been well-tested with puppy (even built for puppy preferably). There should even be some windows programs. So when someone installs puppy they will then have a choice of several good browsers, all the main players plus some more experimental programs, and they all have been tested to work most smoothly and efficiently with puppy. The same with office programs--OpenOffice, K-Office, abiword, and some others. People could configure puppy to work the way they wanted by choosing the programs they wanted and they could mix and match from X, gtk, kde, qt, and even windows programs, but all running in puppy. This is all pretty much the case in current puppy, except that one has to search all the programs out and troubleshoot the things that don't work or are incompatible. And this should also allow the puppyians who are interested to see and compare some good software, and refine our ideas of software and interface design.
But how to get from here to there? Ah-ha, this is the idea I just had--I can start small and do a kind of "proof of concept" test of my own. That is, after 4.1 comes out, I can collect some of the pets I have made and some of the great ones that are in the forums and test them together and place them together on a site with a bit of explanation and see how that works. I know that other people are already doing similar things, but not exactly the same thing I don't believe. I love checking out new software and I can see that many members of the forum do also, so that part would be great fun. This is in the spirit of puppy to me--no need to convince everyone to go along, just do it yourself and make it available to people. Okay, so what is the most important thing to do first? Come up with a cool name of course!
- playdayz's blog
- Login or register to post comments
Technosaurus WebPup
This sounds very interesting--you're right that it is somehow the opposite of my inclinations--but I am looking forward to seeing it--I have already seen the webapps page. Maybe we'll be ready to go at about the same time. thanks.
webpup
I had similar thoughts, but decided to go a whole different direction and remove any programs that have a web app alternative - similar to the google themed "GOS" but somehow I thought "POS" wouldn't go over too well so I have been working on what I am currently calling WebPup. I have already tried a proof of concept using only X and Seamonkey (no ice or jwm) and was able to start most puppy programs using the associated MIME types or creating new ones - for instance associating the puppy podcast grabber with .RSS and creating an empty file called start.rss in order to start it when you click on the link in seamonkey. To supplement this I have been working on the Puppy Web Desktop (pupweb.org/desktop) as a start page to acces all the useful web apps I come across. My goal is a 50MB business card size iso that maintains all of the functionality of Puppy
I am waiting on Dingo 4.1 unleashed and hopefully getting GCC 4.3+ or the 4.2 GEODE patch and associate XORG drivers to work with it since many of the currently available systems that could benefit from this use a geode
Hey, I just wanted to say
Hey, I just wanted to say that I like what you're saying about where Puppy should be going. You've even inspired me to write up some of my ideas. I hope all goes well as you create your proof-of-concept.
my blog posts on Puppy
Proof of Concept
I too am inspired by these ideas. But you describe something I am already familiar with - distros backed by huge software repositories. There is little unique to this.
What I must ask then, how in this vision of yours, is Puppy different? How will it remain different enough to attract the special attention we all wish for it? What qualities will continue to make Puppy unique and more desirable than so many others? How will Puppy be "special" in 2009, 2010, 2011..?
The challenge is this. Today's Puppy is unique because it's creator is unique. Barry is not a community or a committee. He's is one man with ideas and a vision aided by a loyal, creative following. He is a leader. How do we replace this man? How do we as a group re-create the qualities of the individual who grew Puppy to what it is today?
I ask this specific question because it is the most difficult to answer. If we can address it, we are on our way to defining the concept that will define Puppy's future.
unique yes
hi dogone, thanks for the comments, they help clarify the ideas, i do see great uniqueness but i maybe did not communciate it.
One unique thing I see is that the core of the project is puppy with all of its current strengths, fast, lean, install in different ways, install on variety of hardware. My experience is that the big distros (mandriva, ubuntu, etc.) are really pokey because they put in everything anyone might ever need--I mean services and processes running all of the time--they slow it down considerably. Future Puppy will not be like that--it will still be lean and fast--Barry's pup_event replacement for hal and dbus is a case in point. dbus is the one that freaked me out--I created a gnome desktop without it in arch or gentto I can't remember and it was snappy, but when i added dbus it slowed down horribly. Perhaps people will have the option of adding the services that they need but only the ones they need and only if they need them--does that make sense--Puppy will not installl all of them by default. I am one of the ones who favor retaining puppy's uniqueness by continuing to build the system software ourselves, rather than by seeking binary compatibility with a major distro.
xfce may be an example of what i would not want--what the major distros all seem to have--I have been observing if for maybe 5 years--perhaps it is "lighter" than kde or gnome in some ways, but in my experience that lightness definitely *does not translate into speed or responsiveness*--it uses dbus (in fact, i have seen kde run quicker because it uses its massive amounts of memory efficiently at least ;-) But in my experience none of the major desktops that have a modern look and feel are in the same league as puppy--that is what has to be retained..
A second unique thing would be the software repository. I don't really foresee it as huge. Rather, my interest is in software of the highest quality (i know that is nebulous) in terms of function and usability. I mean software that would complement Puppy's virtues of speed and friendliness. I see that the software would be tested to work well with puppy and I think maybe this is the important part, it would be in the repo becuase someone at puppy cared enough to build it and test it and then the group who maintained the repo would also test it. I don't think it would contain every program ever written for linux. The office suites, the cd burners, audio and video players, basically an extension of the current idea of the necessary core software, but with two or three examples of the more elaborate programs--rather than as currently one example of an elegant small program.
I have a hunch that if one went through the Additional Software forum and picked the programs that have gotten some interest from the puppy community--not the ones that were done just because they were easy to do ;-) but the ones that found an audience, that the number of those would not be huge, rather more manageable.
I think it would also be unique the way future puppy would mix and match X programs, kde, qt, gtk, and gnome programs, and wine programs. I think it possible that a kde program group, or gtk, etc., would find an "enabler" base that would allow other programs to be installed more simply. Well, gtk is pretty darn easy!
For many people the base install of future puppy might be all they need, just as now; someone else might need just an office suite to be all set up; someone else might need an audio studio. The future puppy repository could meet the needs of each of those people without them having to chase down dependencies. It is not a small project, but with some groups working on each category of program, and building up to it, not overextending, I think it would be doable.
sorry to go on and on...