What next?

George Makrydakis george at obsethryl.eu
Thu Feb 28 13:59:01 PST 2008


On Thursday 28 February 2008 22:33:09 Ag. D. Hatzimanikas wrote:
> On Thu, Feb 28, at 02:35 George Makrydakis wrote:
> [...]
>
> > 7) Perhaps some of the people majorly involved in this have understood
> > that it is time to stop behaving the way they did. If that is the case,
> > things should fare out well.
> > If not and this is another PR stunt ...
>
> It isn't and really you doesn't have to think this way.

First of all, statistically speaking, most LFS mailing list announcements for 
ideas and implementations tend to create chaotic behaviour among developers, 
contributors and in general, all involved parties. This is the social problem 
I am referring to about. Everyone likes to hear the sound of their own voice 
as a friend said. There is one problem. Mailing lists are a medium of non 
vocal communication, lacking both the emotional intensity (for those who have 
that) that adds particular attributes to meaningful communication as to what 
level of importance certain things are said. This is why I prefer to be 
literal here. And literal I am. If some people take my comments as rants it 
is because they process the information given through an emotional filter I 
am not applying.

> I wrote it in the past and I will repeat it, (for me) its not about who has
> the biggest d* in the market and quite frankly, in LFS we shouldn't care
> for our popularity and we doesn't have to be involved in a comparison game
> with other distributions.
> We are here for a reason (education) and this make us unique.

First of all. It is not only about who has the biggest "name" or "features" in 
the market. It is also about being viable. It is clearly that Gerard's 
initiative on killing LFS the way it is right now will have profound impact 
on the educational aspects. The only educational aspect of LFS and like 
projects has to do with building a proper toolchain for the proper toolchain 
is the hardest and most important thing to do. There should be more "ink" in 
these chapters.

The pythagorian apophthegm "Not everything can be explained to everyone" 
applies to many chapters in the book(s). I will elaborate on why this 
statement can be proved wrong if specialized "streams" of the book can be 
generated for the interested parties, with different skills and requirements.

Take note that if you always try to be simple, you risk taking out complexity 
where complexity is there for a reason. Not to undermine simplicity, for 
simple things are not also meant to be easy... always.

> But, what we have to care is our product. And in this matter you are
> absolutely correct about the superiority of Greg's build method.
> And its not a secret that DIY reference guide had become upstream
> for us in toolchain issues and of course this should be become official
> and start a cooperation with him.
>

I think that if he is reading this email, he is likely to be glad about it. I 
take no stance on who is *better*, right now clfs and diy are simply superior 
in features and technology, overall. The fact that they are, prompts these 
changes.

> Some things I like to mention:  (I am not referring to you particularly)
>
> a. While we are speaking about package management, for me the reason I care
>    mostly about (package management) is only for LSB compliance.
>
>    And while we are talking about standards, another goal we have to reach
>    is FHS compliance and to document any NO POSIX behaviors (mostly
> bash'isms) in packages.

First of all, if somebody wants to have a FHS, LSB compliant do-it-yourself 
system, he can grab the nALFS profiles from the LSB group and build his own 
rpm implementation of a mini LSB distribution. It is not that hard, really. 
The fact that it is not advertised is another matter altogether. LFS and 
others may fall short on some aspects when compared to such an end result, 
but the difference that LFS has to cover is several orders of magnitude 
greater than the one of its competitors.

> b. While the truth is that the project indeed is in maintenance
>    mode, please keep in mind that this is not always an easy task, in fact
>    usually is harder than innovating, because is (some kind of) a boring
>    task.
>    So based on this remark, whatever it will be decided at the end, please
>    try to make future tasks (like to update a package) easier and to
>    encourage wide participation.

Technically speaking, there is a structural problem with LFS's book formats. 
The same is inherited to the now fully self - hosting projects clfs and 
diy-linux. Lack of flexibility. There are lots of ways that this can be 
avoided, extended, changed, evolved etc.

>   [parenthesis] Speaking about participation.
>   The strongest part of LFS is its userland. Everybody alone is a
>   distribution maintainer and precious source of information because this
>   is the nature of LFS way. We carry the exciting burden to build and
>   maintain our personal distribution.
>   So please participate in the development of LFS, not only by writing
>   in the mailing lists (which is our salt/sugar and it is a great tool
>   and a geeky lovely game), but and to write a patch for your favorite
>   program and commit it either to Track or in the {B}LFS mailing list.
>   Participation in Track tickets is also valuable.[/parenthesis]

I have chosen to not partecipate at all in the LFS development and this is why 
I remain a silent observer of this mailing list. The reason has to do with 
the fact that I do not particularly enjoy the way this project is run by a 
particular people who act as public opinion manipulators and strife 
engineers. Whether they do it on purpose or not, it is something that is 
trivial, for repetitive exercise of this policy tends to give a solid trend 
towards what the truth is.

> c. I was involved in such discussions in the past and I had high hopes
>    that things would change. They didn't. What left to me is that usually
>    nothing happens, so please take start to take actions.
>    Here are some tasks: (Mostly to Jeremy)

Interesting you are choosing that particular person. Very interesting.

>    * Start a thread about the name of the new project.
>    * After that; Create a mailing list for the project and use that list
>    from now and on.
>    * Create a Repo.
>    * Start a draft for a manifesto.
>    * Record the ideas from these threads.

People like to receive credit for what is their intellectual offspring, this 
has nothing to do with intellectual property (which imvho is a logical 
fallacy by itself). This does not happen, many times with LFS. I am not 
referring to a particular case, but there are several cases with several 
people. The Absolute Truth Holder Syndrome affected people will have no doubt 
reason to argue on this, but in anycase; it stands.


> d. Speaking about the latter.
>    Many people had offered ideas, plans and scripts. Please -and I am
>    speaking to them mostly - its precious material and it has to be
>    evaluated.

Evaluated yes. Cannibalized no.

>    Currently there is a ton of information that much will be 
>    lost and it is a pity. Please continue to contribute and upload
>    somewhere your scripts/drafts and post them in the new mailing list.
>    In the same spirit keep in touch because we have to fight :) (at least
>    I will) for to make this new project more easy to
>    contribute/participate and quite possible there will be many interest
>    things/tasks to do and I am sure every one will find the desired task
>    to get involved with one way or another.

It will be lost not because there is not a medium for storing this information 
(the mailing list can be read in its archives as well). Simply put, 
popularity contests create enough entropic information to cover meaningful 
information, no matter how important or how trivial.

>    [parenthesis] What the great thing would be if we could attract
>    upstream developers to participate in our project. We are offering by
>    tradition (libraries and binaries) in their purest form, I don't
>    think there is another distribution that is using less patches than
>    us. So I think it worths advertising that LFS is a great platform to
>    test their programs.[/parenthesis]

I understand what you mean.

> e. Speaking about advertisement; we need it. And we need the right people
>    for the job (like Alan for example), people with the right social skills
>    anyway (NOT YOU GEORGE!!!).

You do not know me enough in real life or inet life to be able to judge about 
my social skills or the lack thereof. What I do in real life does employ a 
great deal of patience and understanding I care not to discuss about since I 
consider it non significant information in here. Not in one case have I been 
rude or inconsiderate to anyone even when shamelessly provoked and frankly, 
offended. This has to do with the lack of proper culture and good manners 
from people who tend to not think before they act. You cannot come to terms 
with people who do not deploy logic as a means of effective and objective 
communication. This does not mean that you do not have to protect yourself 
within the limits of decency they do not care to uphold.

Commenting about someone you do not know is definitely the right way to mental 
oblivion. Fortunately, the necessary minority of people understand this 
before it is too late for them, don't they.

>    [parenthesis AND JUST FOR GEORGE] Speaking about you and your social
>    skills is a shame that you don't speak *only* about technical issues.
>    You are obviously a high IQ'er. Please and just for me, because I am
>    the only ellinas in here, make me me a favor and concentrate how to
>    resurrect this project and I am sure you will find many reasons to
>    stay in front of your screen and your editor to contribute some code
>    for LFS.[/parenthesis]

When I spoke of technical issues, I took two precautions: First: POC code (of 
dubious quality to my eyes, but in anycase) written in completely compatible 
with both the level and the minimal functionality expected. People are too 
fast in producing alternatives that touch subjects superficially and they do 
it in such a way that it ends being an unproductive static.

The code is not LFS only now; it is *.

Production code is to be ... different, now.
> f. I am waiting.

Saturday. This one. Perhaps even sooner, but I have to work at work :)

>
>   \ /
>
> > I will limit myself to submitting a proposal for refining your community
> > first, if I may. A technical proposal is to follow in anycase, some
> > people from clfs know about it. Two days left, really.
> >
> > It has been a long time, hasn't it...
>
> Yes, lets see what you have to say.

More than could be of interest here given the overall targets of this 
community. This is why flexibility is at the center of it.

People to whom this is of vital importance, and have approached me have been 
given several bits of information. For what it may concern people who are not 
informed, everything has been redesigned and the approach has been 
experimented with in various settings, in all three (lfs, diy-linux, clfs) 
projects, in both design and implementation. It would have been ready quite 
some time ago, but up until recently, it was a study, put aside, rework 
project.

Not anymore.

Suffice it to say, that as I have stated previously on this mailing list I am 
focused on abstracting everything. This requires study of the viable 
solutions in both meta - distribution and vendor - oriented distribution 
engineering. It was not an easy task and the initial release is to have the 
basic blocks for letting the individual projects implement what they wish 
with it.


> To zoumi kai oxi ta parapona gia tin atimi zoi i gia tous neoyrkezous.
> Na kanoume kati; oi prakseis mas menoun s ayton ton kosmo.
> Arxisa computing sta geramata (sta 37 - stis 23 flebari egina 41)
> giati exasa ena koritsaki kai den exo katholou xrono gia xasimo.
> Eisai ekseritika efyis. Parakalo doylepse gia to project kai stamata na
> kritikareis ton kosmo. San xari sro zitao.

It is difficult to comment something that the rest of the readers will not 
understand because it is written in Greek. Suffice it to say, that my 
criticism towards certain people in this project has to do with the lack of 
respect towards the work of others. This is a fact that has also been 
explicitely reported by several contributing parties and projects, therefore 
it is not just an assumption of yours truly, but a wider, corroborated by 
several pieces of evidence, observation. Plagiarism (in a variable scale) is 
supposed to attract criticism, and it is a cross - cultural observation. 
Perhaps some things are misunderstandings, but their repeated occurrences 
make this scenario unlikely.

For the rest, I sympathize on what you disclosed regarding your personal 
profile and understand.

> > Dux Vitae Ratio
>
> How the hell you managed to speak with Ciaran. Oh I forgot. He lacks
> social skills too.

Since you have read the articles on the website, you will also know that it 
involves several other people as well, on different articles, with more 
things to come provided my time and their time are compatible for producing 
new material.

When you get to reconcile a lot of different mindsets into single piece, you 
put up such an effort that requires several social skills, knowledge and 
ability to bypass what afflicts everyday conversation: unreasonable 
assumptions based on word of mouth that bears little resemblance to the 
truth.

Quite social as a website, isn't it.

Draw your own conclusions as to what I am referring to respect to what you 
incorrectly (perhaps haste is a bad advisor...) assumed as a basic 
personality trait.  Social Engineering is used in this list quite often, I 
have no reason to be part of it. Social skills and social engineering are two 
different things. Some people do not understand the difference. It is their 
call.

> --
> http://wiki.linuxfromscratch.org/blfs/wiki/Hacking





More information about the lfs-dev mailing list