Puppy's big problem with woof and woof CE
-
ITSMERSH
Just checked the current "init" in woof-ce, support for pupmode=77 (Multisession CD/DVD) is still there.tallboy wrote:Earlier Puppys didn't need any storage space, because you could save your session to the multisession CD that contained the .iso, BUT THAT OPTION WAS DELETED SOME TIME AGO IN THE WOOF CEs!
But I seem to remember there being some doubt about the switch to a new CD/DVD when the current one is full, working.
gyro
This assessment is for Fatdog, but I believe Puppy will perform similarly, if not better.Anyway, maybe this is going off-topic, except I feel that it would be nice to have small distro version of Puppy (not using big distro apps etc) so that it can be run in older low RAM machines - but maybe that is the past.
Firstly, as far as I know, no 64-bit machines comes with less than 1GB of RAM, however I set my sights lower and use a hypothetical machine with 384MB with 512MB disk (harddisk, not USB flash drive).
I remaster FD with nano-initrd (kernel-modules and basesfs outside initd - similar to zdrv and pup.sfs outside of initrd in standard Puppy). These files aren't loaded into RAM (similar to pfix=nocopy).
I booted the remaster. I used gparted to partition that 512MB disk into two:
a) sda1 is 256MB swap
b) sda2 is 256MB savedir
Then I activated the swap on sda1 (swapon /dev/sda1).
I launched seamonkey (=known to be a memory hog and not exactly the swiftest browser).
I watched a movie trailer in youtube. It worked well enough. Of course there are delays, but it worked well enough for me to actually be able to watch it. In full screen.
I shutdown the machine, and then reduced the RAM to 256MB, and boot it again. I could still do the same thing, though the video became choppy at the beginning and the webpage wasn't responsive as it was being loaded; and there was noticeable delays between mouse-clicks to the action.
Now, for the record, I tested in qemu. The "harddisk" performance in qemu is probably better than the real ones; and the CPU is one of the recent ones, so I can't really claim on the responsiveness performance; however, the point is to show what is possible with modern Puppies on a very RAM-limited systems - it still can run and work.
That being said. Years ago I attempted to run Puppy 3.0.1 on 128MB Pentium 3 machine (or was it Pentium 2, can't remember). I can't remember the size of the harddisk but it was certainly not big, however big enough for me to create a 64MB (or was it 128MB) swap file. How did it perform? The word "horrible" came to mind. It was too slow to be unusable, delays with every single action. So the "good old days" isn't always as good as we remember it.
Anyway. Small OS is always interesting and has their place. But I believe they fit a very small niche nowadays, because these days small almost always means "very specific". Slitaz for example is a very polished small distro (I've looked at its build system and it does some very interesting tricks to squeeze the very last byte), however it's "key" application is a web browser. That's fine if all you want to do is browsing and using online apps. But once you need to do something else, then you start installing the packages and the size grows.
Fatdog64 forum links: [url=http://murga-linux.com/puppy/viewtopic.php?t=117546]Latest version[/url] | [url=https://cutt.ly/ke8sn5H]Contributed packages[/url] | [url=https://cutt.ly/se8scrb]ISO builder[/url]
I don't really care too much about the size distribution is taking up on my hard drive (despite my machine being from around 2008, still it has 2GB RAM and an old 80GB SSD). With XenialDog64 I basically use a remastered/swollen 01-filesystem.squashfs of size 1GB, but owing to me continually trying out programs and dev systems, it is currently taking up 2.5GB on my HD, so hardly tiny...
I also have the latest Pups on my drive, including most recent DPupBuster for testing, and also FatDog-800 final, which takes very little room on my HD (around 415MB) which is exemplary considering it includes Libre Office, gimp and more.
So yes, very small distro's are a special niche - nice to play with, and Slitaz certainly does a great job extracting every drop of utility out of the Midori version it includes.
My 80GB SSD Hard Drive certainly proves too small (I'm continually running out of space on all the partitions I've created on it. Of course I could fork out and buy a bigger drive but don't feel the need for such an 'extreme' measure. Rather, I'm currently offloading some of the junk I've accumulated (I no longer remember what much of it even is...) onto my external 1TB usb hard drive. That affords me the space to try out other distros again, and maybe even update and re-test makepup to cope with the new distros appearing on woof-CE github site (I do hope newer distro that are being woof-CE built will have their configs uploaded to woof-CE github soon though because, currently, it is a bit of a pain re-configuring makepup to know about new ones).
wiak
I also have the latest Pups on my drive, including most recent DPupBuster for testing, and also FatDog-800 final, which takes very little room on my HD (around 415MB) which is exemplary considering it includes Libre Office, gimp and more.
So yes, very small distro's are a special niche - nice to play with, and Slitaz certainly does a great job extracting every drop of utility out of the Midori version it includes.
My 80GB SSD Hard Drive certainly proves too small (I'm continually running out of space on all the partitions I've created on it. Of course I could fork out and buy a bigger drive but don't feel the need for such an 'extreme' measure. Rather, I'm currently offloading some of the junk I've accumulated (I no longer remember what much of it even is...) onto my external 1TB usb hard drive. That affords me the space to try out other distros again, and maybe even update and re-test makepup to cope with the new distros appearing on woof-CE github site (I do hope newer distro that are being woof-CE built will have their configs uploaded to woof-CE github soon though because, currently, it is a bit of a pain re-configuring makepup to know about new ones).
wiak
Right now I use my 'big and fast' Linuxbox from 2003:So how much RAM do you have on the main machine you are loading modern Puppy into? And how old is the machine/fast is the hard drive?
HP Compaq d530
Motherboard: Hewlett-Packard 0864h
Intel(R) Pentium(R) 4 CPU 2.60GHz
Max Speed: 3200 MHz
Current Speed of Core 0:2593 MHz
Memory Allocation:
Total RAM: 1002 MB
Used RAM: 478 MB
Free RAM: 524 MB
Buffers: 22 MB
Cached: 328 MB
Total Swap: 6159 MB
Free Swap: 4834 MB
Actual Used RAM: 128 MB Used - (buffers + cached)
Actual Free RAM: 874 MB Free + (buffers + cached)
Linux Kernel: 2.6.33.2 (i686)
PAE Enabled: No
Distro: Lucid 5.2.8.7
HDD 40 Gb
I had to get rid of my P2 and P3 PCs, but I have several P4s from the '90s.
True freedom is a live Puppy on a multisession CD/DVD.
-
nosystemdthanks
- Posts: 703
- Joined: Thu 03 May 2018, 16:13
- Contact:
no, it is still doable.wiak wrote:Anyway, maybe this is going off-topic, except I feel that it would be nice to have small distro version of Puppy (not using big distro apps etc) so that it can be run in older low RAM machines - but maybe that is the past.
you need to be able to find the stuff you want to remove and remove it. you could post a list of things to remove.
it gets tiresome to do this by hand, and people know its easy to just remaster-- then they do it a few times, then they give up.
its easier to automate that. then instead of getting tired of remastering, you just run the script and it makes the distro for you.
but you can still do that by hand. what kind of specs are we talking about?
you can even remaster older versions of puppy, if the new ones are too hungry for resources. though i prefer to remaster ones that are up-to-date.
[color=green]The freedom to NOT run the software, to be free to avoid vendor lock-in through appropriate modularization/encapsulation and minimized dependencies; meaning any free software can be replaced with a user’s preferred alternatives.[/color]
If you write your own script then you can make anything within your potential coding ability. I unfortunately haven't tried woof yet but it seems to me that you could use the packages within woof for whatever kind of system that you want to build...or at least you could if said packages are part of the system that you want.wanderer wrote:is there no way to just build a minimal woof-ce
with a script
just a base
that you could add to later
1 console only
2. basic x
3 apps you like
it could build fast
and be simple enough for non gurus to understand
then that could be the community edition woof-ce mini puppy
just wondering
wanderer
The woofCE code has a base skeleton for a puppylinux system,
woof-CE/woof-code/rootfs-skeleton/
and a bunch of standard puppy packages that are related to that base skeleton.
woof-CE/woof-code/rootfs-packages/
Many of these packages have a standard install script. Simply copy the files within these packages into your target environment and then run the post install script.
I'm sure there is more to Woof then simply a skeleton and a bunch of packages but this alone is a good starting point for a distribution.
Anyway, I have written code to do what I describe but it isn't quite ready.
If you understand the build system then you can modify the build system while you are tweaking the OS, this way if you mess up a tweak you just run the build system again to hopefully get close to where you left off. Backups of course help.nosystemdthanks wrote:no, it is still doable.wiak wrote:Anyway, maybe this is going off-topic, except I feel that it would be nice to have small distro version of Puppy (not using big distro apps etc) so that it can be run in older low RAM machines - but maybe that is the past.
you need to be able to find the stuff you want to remove and remove it. you could post a list of things to remove.
it gets tiresome to do this by hand, and people know its easy to just remaster-- then they do it a few times, then they give up.
its easier to automate that. then instead of getting tired of remastering, you just run the script and it makes the distro for you.
but you can still do that by hand. what kind of specs are we talking about?
you can even remaster older versions of puppy, if the new ones are too hungry for resources. though i prefer to remaster ones that are up-to-date.
The point of the build system is that it's reproducible whereas one will forget the steps that they did by hand.
the problem for me
and i think a lot of puppy users
is that we don't have the technical ability
to write scripts of that sophistication
we could use one if it was fast and simple
and this would result in puppy having a minimal version
that people could develop
i think it would be a great asset to the puppy community
but until a guru with the required skill volunteers to do it
i will just work on corepup
which is a system that does the same job
wanderer
and i think a lot of puppy users
is that we don't have the technical ability
to write scripts of that sophistication
we could use one if it was fast and simple
and this would result in puppy having a minimal version
that people could develop
i think it would be a great asset to the puppy community
but until a guru with the required skill volunteers to do it
i will just work on corepup
which is a system that does the same job
wanderer
woof & woof CE are too Powerful
In creating Woof, Barry K developed a means by which Puppy could survive without his constant input. For that, we are grateful.
But Woof is too powerful. As envisioned by Barry woof could be used to create a Puppy using (almost) any distro's binaries. We have taken advantage of that possibility, creating slackos, dpups, upups, Fatdog (woofing Linux-from-scratch), an archpup, and IIRC a Puppy based on Mageia. There are 32-bit Puppies, and 64-bit Puppies and some FatDogs supporting the Arm-architecture. And remasters of many. I think Ally indicated at last count there was around 5000 Puppies (woof-builds, remasters, pre-woof, and hybrids).
The three major suppliers of binaries used in Puppys are Slackware, debian and Ubuntu. Every time one of those makes a significant change the thousands of lines of code which make up Woof in its divisions must be examined and, if necessary, edited. Maintaining Woof is a Full-Time job for which no one is paid and few even receive recognition.
While Puppy has many fans with good ideas [and some not so good] what it lacks is sufficient expert programmers to keep Woof up-to-date with independent double-checking of coding and responding to feed-back of from the fans as to how the woof-product functions in a real world environment.
Currently, Slackware (and/or its forks) maintain repositories for five versions: Slackware-Current, 13.37, 14.0, 14.1 and 14.2. Debian has a testing branch (currently Buster) and a Stable branch (currently Stretch) as well as an Unstable-branch. Perhaps Jesse is an 'old-stable' branch. At any rate, while there are repos for the three named, there are also repos for debian sid and debian wheezy. debian expects that volunteers will continue to provide support for Jessie thru June 2020, stretch thru June 2022. I would expect that support will continue for Buster thru 2024. Currently, Ubuntu maintains Trusty Tahr, Xenial Xerus, Bionic Beaver and Cosmic Cuttlefish whose End of Life in June of this year will occur two months after its release of another short-term release, disco dingo.
Ubuntu uses these short term releases as test-beds for new ideas which may, or may not be incorporated into its next long-term release or backported to a still maintained long term release.
And as I mentioned before, every significant change in one of those 'up-stream' distros has an impact on the maintaining woof.
Do you remember when there was one Official Puppy, published by Barry K and every other Puppy was a remaster? The ability of anyone to remaster a Puppy has long been one of its most attractive features even when it was a time-consuming process requiring concentration and user input. Thanks to shinobar and nic007, much of the drudgery and guess-work has been taken out of remastering.
Having to produce an entire distribution to test new ideas or include kernel changes is also not 'the Puppy way'. Thanks to jemimah's pioneering work in modularization even a newbie can quickly replace kernels and their associated drivers. Thanks to sailor enceladus, that can now be easily accomplished by newies in a Slacko 5.7 based on Slackware first published in 2012. Easy kernel upgrades have always be the case with later Slackos, and 'Ubuntus' since 2014. [It may be possible with earlier puppies, but with some difficulty and perhaps only by the creation of a 'frankenpuppy'].
Einstein is quoted as saying that "Genius is 1% inspiration and 99% perspiration". We are all gifted with two mental systems --the gestalt and the analytical. The gestalt takes in the whole world in a glance, evaluates it against experience, see the change or the possibility of change. It is the 'Eureka' moment; the race-car driver 'in the zone'; the athlete's instant response to stimulus, the artist's inspiration. The analytic is a language function: to see the parts, provide names for them, to see the relationship of one part to another and provide names for that relationship, and then to mentally visualize altering the parts and/or their relationship. Sometimes the Eureka moment can only arise after a great deal of seemingly fruitless effort, but an effort by which we obtain a pre-requisite experience. When it occurs we are stimulated to pursue our new understanding through out analytic mental system. Analysis is always hard-work.
I've heard it said that one has not mastered a language until one can dream in it. How many of us can dream in bash, or ash, or dash, or any other programming language? Whatever idea we may have in whatever our native language is, however we visualize the relationship of one thing to another, to effectuate those ideas, those visions in a computer language alone requires employment of the analytical function. I would suggest that regardless of how much any of us may enjoy solving puzzles, there comes a point at which mental exhaustion sets in. Solving a puzzle may bring a sense of relief, even of pride in our ability to have done so. But unlike a Eureka moment, it does not stimulate us to continue. The puzzle is solved, there is nowhere further to go, it is time to rest.
However good we are at solving puzzles if there is no rest, if all we can visualize is that soon there will be just a variant of the last puzzle with nothing really new and challenging to discover, solving puzzles becomes just another unexciting chore.
And here's the thing about our two mental systems. We can't operate them simultaneously. Maintenance of Woof is a chore. Whoever undertakes it as a full time responsibility will not have the opportunity to explore (experience vicariously) other developments taking place in Linux, not have the opportunity to pursue other visions of the possible.
Two of our most creative devs were 01micko and pemasu. Both became involved in maintaining woof. Since then, I recall no further input on any other matter of interest on the Forum. Are they even still with us? Perhaps 'real-world' interests now occupy their time? Or perhaps, just maintaining woof turned Puppy for them into a chore rather than a means of accomplishing something creative.
Barry has mentioned that this has been a year of great creativity on the Forum.
[To be continued]
But Woof is too powerful. As envisioned by Barry woof could be used to create a Puppy using (almost) any distro's binaries. We have taken advantage of that possibility, creating slackos, dpups, upups, Fatdog (woofing Linux-from-scratch), an archpup, and IIRC a Puppy based on Mageia. There are 32-bit Puppies, and 64-bit Puppies and some FatDogs supporting the Arm-architecture. And remasters of many. I think Ally indicated at last count there was around 5000 Puppies (woof-builds, remasters, pre-woof, and hybrids).
The three major suppliers of binaries used in Puppys are Slackware, debian and Ubuntu. Every time one of those makes a significant change the thousands of lines of code which make up Woof in its divisions must be examined and, if necessary, edited. Maintaining Woof is a Full-Time job for which no one is paid and few even receive recognition.
While Puppy has many fans with good ideas [and some not so good] what it lacks is sufficient expert programmers to keep Woof up-to-date with independent double-checking of coding and responding to feed-back of from the fans as to how the woof-product functions in a real world environment.
Currently, Slackware (and/or its forks) maintain repositories for five versions: Slackware-Current, 13.37, 14.0, 14.1 and 14.2. Debian has a testing branch (currently Buster) and a Stable branch (currently Stretch) as well as an Unstable-branch. Perhaps Jesse is an 'old-stable' branch. At any rate, while there are repos for the three named, there are also repos for debian sid and debian wheezy. debian expects that volunteers will continue to provide support for Jessie thru June 2020, stretch thru June 2022. I would expect that support will continue for Buster thru 2024. Currently, Ubuntu maintains Trusty Tahr, Xenial Xerus, Bionic Beaver and Cosmic Cuttlefish whose End of Life in June of this year will occur two months after its release of another short-term release, disco dingo.
Ubuntu uses these short term releases as test-beds for new ideas which may, or may not be incorporated into its next long-term release or backported to a still maintained long term release.
And as I mentioned before, every significant change in one of those 'up-stream' distros has an impact on the maintaining woof.
Do you remember when there was one Official Puppy, published by Barry K and every other Puppy was a remaster? The ability of anyone to remaster a Puppy has long been one of its most attractive features even when it was a time-consuming process requiring concentration and user input. Thanks to shinobar and nic007, much of the drudgery and guess-work has been taken out of remastering.
Having to produce an entire distribution to test new ideas or include kernel changes is also not 'the Puppy way'. Thanks to jemimah's pioneering work in modularization even a newbie can quickly replace kernels and their associated drivers. Thanks to sailor enceladus, that can now be easily accomplished by newies in a Slacko 5.7 based on Slackware first published in 2012. Easy kernel upgrades have always be the case with later Slackos, and 'Ubuntus' since 2014. [It may be possible with earlier puppies, but with some difficulty and perhaps only by the creation of a 'frankenpuppy'].
Einstein is quoted as saying that "Genius is 1% inspiration and 99% perspiration". We are all gifted with two mental systems --the gestalt and the analytical. The gestalt takes in the whole world in a glance, evaluates it against experience, see the change or the possibility of change. It is the 'Eureka' moment; the race-car driver 'in the zone'; the athlete's instant response to stimulus, the artist's inspiration. The analytic is a language function: to see the parts, provide names for them, to see the relationship of one part to another and provide names for that relationship, and then to mentally visualize altering the parts and/or their relationship. Sometimes the Eureka moment can only arise after a great deal of seemingly fruitless effort, but an effort by which we obtain a pre-requisite experience. When it occurs we are stimulated to pursue our new understanding through out analytic mental system. Analysis is always hard-work.
I've heard it said that one has not mastered a language until one can dream in it. How many of us can dream in bash, or ash, or dash, or any other programming language? Whatever idea we may have in whatever our native language is, however we visualize the relationship of one thing to another, to effectuate those ideas, those visions in a computer language alone requires employment of the analytical function. I would suggest that regardless of how much any of us may enjoy solving puzzles, there comes a point at which mental exhaustion sets in. Solving a puzzle may bring a sense of relief, even of pride in our ability to have done so. But unlike a Eureka moment, it does not stimulate us to continue. The puzzle is solved, there is nowhere further to go, it is time to rest.
However good we are at solving puzzles if there is no rest, if all we can visualize is that soon there will be just a variant of the last puzzle with nothing really new and challenging to discover, solving puzzles becomes just another unexciting chore.
And here's the thing about our two mental systems. We can't operate them simultaneously. Maintenance of Woof is a chore. Whoever undertakes it as a full time responsibility will not have the opportunity to explore (experience vicariously) other developments taking place in Linux, not have the opportunity to pursue other visions of the possible.
Two of our most creative devs were 01micko and pemasu. Both became involved in maintaining woof. Since then, I recall no further input on any other matter of interest on the Forum. Are they even still with us? Perhaps 'real-world' interests now occupy their time? Or perhaps, just maintaining woof turned Puppy for them into a chore rather than a means of accomplishing something creative.
Barry has mentioned that this has been a year of great creativity on the Forum.
[To be continued]
Oooh, that was deep, Mike! And absolutely brilliant.
wanderer, you have been around for a while, so you should remember the choice you had, when downloading an early Puppy. (like 2.17?) You could use the OOTB version, or build your own by adding programs to a base. I cannot find the old repos, and I cannot remember the correct names for the processes.
What I would like to see, is a package that could compile any source, for use in the Puppy you currently use. The compile process is too complicated for me to use, as it is. The dream is to be able to just drag a source file to a desktop icon, perhaps named 'Adapt', and voilá!
wanderer, you have been around for a while, so you should remember the choice you had, when downloading an early Puppy. (like 2.17?) You could use the OOTB version, or build your own by adding programs to a base. I cannot find the old repos, and I cannot remember the correct names for the processes.
What I would like to see, is a package that could compile any source, for use in the Puppy you currently use. The compile process is too complicated for me to use, as it is. The dream is to be able to just drag a source file to a desktop icon, perhaps named 'Adapt', and voilá!
True freedom is a live Puppy on a multisession CD/DVD.
hi mikeslr and tallboy
i agree
so this is an open letter to the puppy gurus
you gurus have done a superhuman job
but you are the victim of your own virtues
we all really appreciate all you have done
and are truly in awe
your genius and untiring work
has created something
that is a masterpiece
but
it cannot be used or maintained
by the average person
this is not a problem for me
because i have always been a minimalist
so woof or woof-ce
was never really an option for me
i did run them a few times
but they weren't the way i wanted to do things
i always was playing with systems
like micromuppy and damn small linux
now i am playing with corepup/tinycore
it is a very good system
i would suggest people have a look
puppy and the puppy community
will do fine no matter what
as long as there is a forum to get together on
but i do think a small simple "official" puppy
would be nice as a starting point
for a lot of people
and would take a lot of the
"if its not woof-ce its not puppy"
arguing away
i would suggest making it into small components
that could be put together as desired
like tinycore
but with sfs files and pets instead of tcz
maybe even look into pupngo as a starting point
pick just one repository
debian stable or whatever
but like i said
nothing really has to be done
as long as there is a forum
there is a puppy community
carry on
wanderer
i agree
so this is an open letter to the puppy gurus
you gurus have done a superhuman job
but you are the victim of your own virtues
we all really appreciate all you have done
and are truly in awe
your genius and untiring work
has created something
that is a masterpiece
but
it cannot be used or maintained
by the average person
this is not a problem for me
because i have always been a minimalist
so woof or woof-ce
was never really an option for me
i did run them a few times
but they weren't the way i wanted to do things
i always was playing with systems
like micromuppy and damn small linux
now i am playing with corepup/tinycore
it is a very good system
i would suggest people have a look
puppy and the puppy community
will do fine no matter what
as long as there is a forum to get together on
but i do think a small simple "official" puppy
would be nice as a starting point
for a lot of people
and would take a lot of the
"if its not woof-ce its not puppy"
arguing away
i would suggest making it into small components
that could be put together as desired
like tinycore
but with sfs files and pets instead of tcz
maybe even look into pupngo as a starting point
pick just one repository
debian stable or whatever
but like i said
nothing really has to be done
as long as there is a forum
there is a puppy community
carry on
wanderer
As a reminder, folks...
If you open a console and type:I hope you won't fall off your chair...
That's the sum of all ro and rw directories in /initrd, in other words the total size of
the Puppy, including perhaps Open|Libre|Office, maybe a java JRE, the devx, etc.
Here is another formula, if you want to see the size of the pup_r?'s individually:
Get the "small" concept out of your heads, people: Puppy is actually a medium-
sized distro packaged to look small. Given its versatility, I frankly see nothing wrong
with Puppy being "medium-sized".
BFN.
If you open a console and type:
Code: Select all
df | awk '$NF ~ /pup/ {sum += $3} END {print (sum/1024)}'That's the sum of all ro and rw directories in /initrd, in other words the total size of
the Puppy, including perhaps Open|Libre|Office, maybe a java JRE, the devx, etc.
Here is another formula, if you want to see the size of the pup_r?'s individually:
Code: Select all
df | awk '$NF ~ /pup/ { print $NF,$3 }'sized distro packaged to look small. Given its versatility, I frankly see nothing wrong
with Puppy being "medium-sized".
BFN.
musher0
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
-
oui
Puppy came as "the" absolutely TINY distro!musher0 wrote:Puppy is actually a medium-sized distro packaged to look small. Given its versatility, I frankly see nothing wrong with Puppy being "medium-sized".
not as medium-sized, that is only YOUR invention because you would welcome rebuild fresh Puppy's but are note able to do that...
the word "DOG" is now occupied by our new neighbours from the dog series...
go away with you big pseudo puppy's and find a new name for them, I would say MIDDLE-SIZED-LINUX-DERIVATED-ANIMAL!
Those numbers I did NOT invent, oui.
Have you tried the awk formulas I gave above on your
Puppy? What results do you get?
Also, if you'd be kind enough to remember that I am NOT
a member of the woof-CE group.
Oh, I almost forgot. If it can improve your mood, give me
your postal address: I'll send you a punching bag with
my picture on it. I'll do anything to help a fellow Puppyist
out of his depression, you know.
BFN.
Have you tried the awk formulas I gave above on your
Puppy? What results do you get?
Also, if you'd be kind enough to remember that I am NOT
a member of the woof-CE group.
Oh, I almost forgot. If it can improve your mood, give me
your postal address: I'll send you a punching bag with
my picture on it. I'll do anything to help a fellow Puppyist
out of his depression, you know.
BFN.
musher0
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
~~~~~~~~~~
"You want it darker? We kill the flame." (L. Cohen)
-
Colonel Panic
- Posts: 2171
- Joined: Sat 16 Sep 2006, 11:09
Just my take on this; when I first downloaded a copy of Puppy in June 2006, it fitted on a single CD-R as about half of Linux distros did in those days. Now, in 2019, Puppy still does although most other distros no longer do. Something to be thankful for IMO.
Gigabyte M68MT-52P motherboard, AMD Athlon II X4 630, 5.8 GB of DDR3 RAM and a 250 GB Hitachi hard drive running Ubuntu 16.04.6, MX-19.2, Peppermint 10, PCLinuxOS 20.02, LXLE 18.04.3, Pardus 19.2, exGENT 200119, Bionic Pup 8.0 and Xenial CE 7.5 XL.
-
oui
Amendment to your friendly message dear Colonel Panic:
credit card sized business cd
or on a
3" = 7,7 cm single CD
(like SliTaz continue to fit on them both today - perhaps for the last time, but the actual version SliTaz 5.0 Rolling continues and it is probably the only one version with 64 bit kernel fitting on those mini-CD's)!
It was possible to get such Puppy's until version WhiteFang (VESA only!), 49,7 MB, really a great (but not all the scope of Puppy applications) relatively modern Puppy for VESA able machines.
Puppy 2.0 with Open Office did have the size 99 MB.
One of the best modern Puppy's is and stay to be Slacko-5.3.3 (recommendable also for machines with big RAM) with a few more than 110 MB). Slacko-5.3.3 is absolutely top! All newcomers with old and small machines have to test it (accede using the micko page and ibiblio)! It continues to be one of the best Puppy's at all but would need a new browser or an extension like SliTaz offers for SliTaz with his package libfirefoxESR and some midori or analog making such old browser able to do what actual surfers often need, as the in Slacko-5.3.3 built in Seamonkey 2.9 has really to much limits and actual versions are not directly able to be use in Slacko-5.3.3 ...). Slacko-5.3.3 offers all the usual Puppy app's.
Quirky did also offer versions with real little size (with of not with all the scope of usual Puppy applications) but separate from the Puppy line.
As you did download your first ISO of Puppy in June 2006, it probably did continue to fit on aColonel Panic wrote:Just my take on this; when I first downloaded a copy of Puppy in June 2006, it fitted on a single CD-R as about half of Linux distros did in those days. Now, in 2019, Puppy still does although most other distros no longer do. Something to be thankful for IMO.
credit card sized business cd
or on a
3" = 7,7 cm single CD
(like SliTaz continue to fit on them both today - perhaps for the last time, but the actual version SliTaz 5.0 Rolling continues and it is probably the only one version with 64 bit kernel fitting on those mini-CD's)!
It was possible to get such Puppy's until version WhiteFang (VESA only!), 49,7 MB, really a great (but not all the scope of Puppy applications) relatively modern Puppy for VESA able machines.
Puppy 2.0 with Open Office did have the size 99 MB.
One of the best modern Puppy's is and stay to be Slacko-5.3.3 (recommendable also for machines with big RAM) with a few more than 110 MB). Slacko-5.3.3 is absolutely top! All newcomers with old and small machines have to test it (accede using the micko page and ibiblio)! It continues to be one of the best Puppy's at all but would need a new browser or an extension like SliTaz offers for SliTaz with his package libfirefoxESR and some midori or analog making such old browser able to do what actual surfers often need, as the in Slacko-5.3.3 built in Seamonkey 2.9 has really to much limits and actual versions are not directly able to be use in Slacko-5.3.3 ...). Slacko-5.3.3 offers all the usual Puppy app's.
Quirky did also offer versions with real little size (with of not with all the scope of usual Puppy applications) but separate from the Puppy line.
That's not what Puppy was known for. It wasn't famous for being the tiniest distro. It was revered as the absolute king of the lightweight distros because Puppy was a complete, fully featured, lightweight, portable, and it "just worked" in all kinds of install scenarios.oui wrote:Puppy came as "the" absolutely TINY distro!
As you probably know, tiny was the realm of Damn Small Linux (aka Tiny Core) or Slitaz for business card CD, or Slax for mini CD. Puppy was small too, but not the smallest. It was, however, fully featured while still being small, and extensible to having the capabilities of a full distro.
I also don't get this obsession with small size lately. Size is relative. Small compared to what? A full size distro? Well, back when a distro barely fit on a single CD, Puppy was under 200MB. And today where distros are around 1.5-2 GB DVD size, Puppy or Fatdog is around 300-400MB. So it's stayed around the same relative size, 20-25% the size of a full distro.
Look, progress marches on. Computers advance.(albeit more slowly now these days) As others have said before, newer computers and technologies demand more modules and libraries and firmware. Sure, you can make a distro small by leaving all that annoying functionality stuff out, but what good is that?
In my mind, it's kind of...how do I say...dishonest? to brag about a 50MB size distro that can't do squat until you add a whole bunch of stuff to it, like Office, graphics, audio and video, full web browser, tools and utilities. Its a bait and switch.
And old is relative too. Old computer is a sliding scale that today the bottom end is a Pentium 4 with less than 1GB ram. That was the high end in the early Puppy days. Sure you can design a distro to work with less than that, but for what? A very small audience and a very painful computing experience, not worth the effort.