[blfs-dev] Mercurial or wget for openjdk?

Fernando de Oliveira famobr at yahoo.com.br
Sun Sep 6 16:35:54 PDT 2015


Ken, thank you very much for the post. You and Bruce are the source of
most of my knowledge and thought most of what i knoe about editing the book.

I will never stop liking you both very very much indeed.

Em 06-09-2015 17:31, Ken Moffat escreveu:
> On Sun, Sep 06, 2015 at 04:15:01PM -0300, Fernando de Oliveira wrote:
>> Em 06-09-2015 10:03, Pierre Labastie escreveu:
>>
>>> I have a proposition, which might satisfy everybody: why not put both the
>>> tarball download and the mercurial instructions in the book?
>>
>> Please don't do that. Forget wget. I don't need to be satisfied, but
>> respected. If you had not "forgotten" my earlier post, no threads about
>> this subject would have happened. And i don't like my idea side by side
>> with one that I completely disagree.
>>
> 
> In much of BLFS there are things that _I_ disagree with.  Many of
> them are a matter of taste, or of my preference for "every package
> in my normal build needs to justify its inclusion" - to me, a
> smaller build is better (but these days, small seems to have fallen
> by the wayside - I increased my system sizes to 14+GB, but they
> still fill up).

This host which is ending second use as development machine:

Sist. Arq.      Tam. Usado Disp. Uso% Montado em
/dev/root        24G   20G  2,9G  88% /
tmpfs           7,9G  852K  7,9G   1% /run
devtmpfs        7,9G     0  7,9G   0% /dev
/dev/sdb3        20G   15G  4,3G  77% /home
/dev/sdb10      9,8G  7,1G  2,2G  77% /opt
/dev/sda2       1,3T  771G  454G  63% /tmp
/dev/sdb6       7,8G  5,6G  1,8G  76% /mnt/LFS72


> 
> On a couple of occasions over the years I have come close to throwing
> my toys out of the (BLFS) pram, but I managed to back off for a day
> or three, and think about what the consequences would be for *me* : a
> lot more time available from not updating BLFS, but a far greater
> expediture of time to find out how to build everything.  By that
> point, I decided that for me, going off in a huff would not bring any
> real benefit, and I was able to look again at whatever the issue was,
> and realise that really it was not important.

Agreed.

> 
>> This a situation without that no one remedy any more.
>>
>> As I am writing this post before than I wishwd to:
>>
>> To all editors:
>>
>> Em 06-09-2015 01:26, Bruce Dubbs escreveu:
>>> Fernando de Oliveira wrote:
>>
>>>> If there is a choice between X and Fernando, X will be chosen.
>>>
>>
>>> That is defiantly not true. Actually I try to not consider the
>>> source of the suggestion, but only the pros and cons of that
>>> suggestion. There are commits that are made all the time where I
>>> would do things differently, but what is there is good too so a
>>> comment or change is not needed.
>>
>> After I fix what I can, I will leave open the tickets to be inspected,
>> changed at will, and please, close them afterwards. This applis to the
>> main BLFS editors, not others.
>>
> 
> Please, nobody is criticising what you do (from time to time there
> might be issues in the detail - that applies to *all* of us).  We
> each have different strengths and weaknesses - I continue to be
> impressed by how many package updates you manage to do, and I was
> impressed this weekend that you found some upstream commit(s) to fix
> Mesa with llvm-3.7.0.

I do accept criticisms, you and I had recent discussions about sddm,
lxqt, and I believe I was not inflexible.

However, sometimes we can say nice things things to someone else
enclosing s subliminal contrary message. This is not your case, thanks
for being so kind, but i can't resist to say that British people are the
best masters in this technique. :-)

> The important thing is to value differences and diversity.  Anybody
> who becomes an editor should be, and is, respected.  That applies to
> those editors who happen to work on systemd equally with those who
> work on trunk.

Agreed.

> We all have different interests, and different beliefs, different
> things which we hold dear, and different prejudices.  Many of us
> have individual problems too.   But none of that should cause us to
> make disrespectful comments about each other.  From my own
> experience, when things get me down I know it can be easy to become
> inward-looking and draw false conclusions.

Me too.

> Sometimes, a brief pointer to what I am doing wrong can feel like a
> slap around the face with a wet fish.  That is not usually how the
> sender intended it to feel.  On technical lists, it is conventional
> to be terse (sorry, _I_ often fail that part).

OK.

> Discussion is good, but for most things there is no vote, somebody
> makes a decision.  The alternative is that we *all* ask, *every*
> time : 'should I do this, or should I do that ?' for a particular
> issue in a package upgrade, then take a vote - that cannot work
> (unless we are only going to update a couple of packages per week),
> and we all rely on all the other editors' judgement.

Yes. But this is not the point here.

1. Before the past update, there was an agreement between Pierre and I
that wget would be used.

2. The update came, but not wget.

3. Then, X expressed a concern similar to mine, but with a different,
never used before in BLFS,solution, involving hg.

4. Bruce, Pierre, nobody talked about 1., and surprisingly, Bruce, who
agrees with whatever Pierre wants(that is the reason I abandoned OJDK),
immediately agreed. (My imagination is working hard the last days about
how many pages I could modify to use git, hg, snb, etc.)

5. I reminded Pierre about 1.

6. Pierre agreeded.

7. Pierred disagreeded.

8. Fernando complaints.

9. Pierre suggests two sets of instructions in the same page.

10. Bruce agrees.

11. I respectfully disgreed and asked to forget my suggestion. Now, less
respectuflyy, it is ridiculous two sets of download instructions just to
satisfy A and B, even B being Fernando.

I could just have done the modification myself, when Pierre "forgot".
But I acted like you, ken. Not a big deal for a fight. But the the same
discussion started again.

And it is not some thing of moment, it started months ago.

>> Also, it was before the bad, unfortunate, "systemd" incident that put
>> Bruce and Fernando in opposite sides. Without that, I would always
>> continue having all the respect, admiration and liking for Bruce.
>>
> 
> Some people like systemd, even among LFS users.  Others have to use
> packages which only work on systemd.  Do not allow your dislike of
> systemd to poison your views or your attitude to other editors and
> developers - on lkml a few months ago there were anti-systemd trolls
> posting, throwing insults (well, what a right-winger who wanted
> women to be subordinate to men would think were insults).  I almost
> concluded they were sock-puppets set up to make systemd look good.
> Please do not fall into that trap.
> 
> ĸen
> 

I am not in that trap.

I wish the systemd book all success, perhaps will be there in the
future, like Douglas very much, and have good impression from him,
except from an incident which drawn a line between Douglas and me. The
dividing line is *not* systemd.

I have 5 working machines running Gnu/Linux, only this host is not
running systemd.

LFS and BLFS have separated ticket systems and e-mail lists. This should
be the same for BLFS and BLFS systemd.

When the systemd branch started, I thought it wouldn't be necessary.
Wrong I was.

Each update, I explicited the systemd switches, which brought me the
necessity to include a Command Explanation. Reason: Facilitate systemd
to just change a not to yes and a not supported to supported. I.e. I did
part of the work trying to help.

Consequence: the editor complained that I should use without. I
immediately stopped worrying about.

Then, I woke up with a surprising complete change of the books, when I
had been agreed that it would'nt be necessary: packages.ent and
gnome.ent are born. I wouldn't be against, but why the necessity to do
it behind all backs? It was a strctural change of the book, not a
misplaced comma.

Then another systemd editor appeared, they fought with each other, split
and another branch is born.

I tried to help the new editor and he started thinking that I was trying
to sabotage.

Both stepped out.

I had helped Douglas, privately, created tickets when he had problems,
gave technical advice, combined that he could take any tickets from me,
everything was perfect.

When the last editor stepped down the systemd book, immediately Douglas
decided to take over, but wanting to be at the same time at BLFS. What?
Aren't the whole systemd LFS and systemd BLFS enough?

I told him I was sorry but it couldn't be possible, due to all the
previous problems with systemd people, I was sorry to have him leaving
BLFS, wished him all the success with systemd.

Asked him to slow down our contacts, because from my previous
experience, there was always a problem between taht systemd branch and BLFS.

Asked him to not interfere with BLFS and I wouldn't try to interfere
with anything of his branch. yes, it is all his.

After that, helped him at least once.

Even told him that he could just copy the updates from BLFS in what
wouldn't be different, make changes as he wished, but pleas *I didn't
want acknowledgement*.

That was a source of problems with the previous systemd editors. The
felt in obligation to acknowledge what they got from BLFS, and started
wishing to "contribute" to BLFS and that I acknowledged that "contribution".

So, it was combined: no intrusion from each other, no need for
acknowledgement.

One day, Duglas started giving contributions to BLFS> tickets, patches,
errors, etc.

I opened a ticket and closed the other as duplicate.

Next day or the same day, I received from the Bruce a boss letter very
kindly asking me to not delete systemd needed things.

Asked Bruce: are you talking about the duplicate ticket?

Answer:

"Perhaps".

After two days of being upset, now weeks being upset, i told him he
should know me better, I don't touch systemd, I try to be maximum
honest, without prejudice and I'm not a kid, next year will be 60. And
from that day on I would never reply to post from him that would
necessitate to defend myself. And would only be going on with BLFS
because I like it and all the people here, including him and Pierre. Of
course this is true with respect to you, too.

I never considered me right-winger (if I understand well this word) and
although systemd is a plague and much probably will be universally
spread, the splitting line is that I don't consider respectful to be
blamed for something I didn't do, and when ask an explicit definition of
the cause of the blame, the boss doesn't actually. No This is not a
serious, respectful attitude.

I'm having difficulty for working, but I'm sure my once acquired
cristian moral will make me forget everything eventually.


The splitting line is not systemd.

-- 
[]s,
Fernando


More information about the blfs-dev mailing list