On 17-10-2014 22:55, Bruce Dubbs wrote:
> Fernando de Oliveira wrote:
>> OK.
>>
>> Apologies to you all.
> And I think a lot of people appreciate that a lot.
I am making a big effort trying to think that, from the set of
(B)LFSers, the subset appreciating what I do is not empty. I am sure
that the subset complement is highly populated. Noticed that Bruce wrote
in third person, which apparently implies in which subset he is.
:-(
>> From now on, I will keep doing my work in updates, as long as I am not
>> having headaches, but won't mind anymore if anybody changes what I have
>> done. It is not worth my mental health, having an opinion and fight for
>> it, so, I will try my best to not give opinions about anything.
>>
>> Armin, from my part, you are free to change all I have done or will do,
>> as you please.
>>
>> Also, it is not worth asking me about something someone thinks I did
>> wrong: my anticipated response, please know, is yes, you are right and I
>> am wrong, go ahead and fix it.
>
> You seem to be tired. I don't want you to do that. Please rest up a
> bit, but consider inputs as suggestions for consideration, not errors.
> We all make mistakes and working together makes us all better.
With this problem in my mind I get more tired each day, without being
able to concentrate in other activities, trying to solve this problem.
I have received a voice communication from Alice, who lives in a Mars
space lab.
She was confused. In her system Freetype-2.5.1 is installed. She had
discussed with her friend Alex who is also in the space lab and
participated in the interplanetary voice communication conference. Has
Freetype-2.5.0 in his system. Both are asking me the meaning of "newer
versions of Freetype", because it looked more like a comment than an
information. Should them both use the sed?
It was a difficult communication, due to the delay between sending and
receiving the signal, she apologized for that. I told her that here on
Earth things are even more difficult, delays are even longer.
After telling her that the book is wrong, and how the original text was,
she thanked me and suggested: why don't they at least write the first
version needing the fix?
I told her that Armin, who shares with me the difficulty of accepting
something different from what we want, passed the new text with another
fix, naming it "improvement" [1], named the patch "meh", which is very
symbolic. Difference between us is that he based his wish on his
preferences and rejection for anything seemingly mathematical, while I
based on whether the sentence carried a non null amount of information.
Using "newer", the amount of information is as close to zero as anybody
reading wants.
BTW, mathematical expressions are all over configuration scripts, all
shells need some operators for comparing numbers, texts, order, sort etc.
And the boss, Bruce, who has much more important things to do, didn't
have enough time to pay enough attention. So, he let it through [2],
without asking the author (me) if that was necessary.
He argued that previous format would need to be eventually updated, form
now in the book, wouldn't. Of course not, as the text in the book
informs nothing! Worse yet, the work of "updating" had anyway been done,
even before such a necessity happened!
He also asked why a user wanted to know exact intial version, but
Alice's questions answers that. And the reason is obvious. Otherwise,
edition and creation of patches (which the sed's could be transformed
into) should be modified, e.g.
{{{
Submitted By: Igor Živković <[email protected]>
Date: 2013-10-08
Initial Package Version: 1.1.7
Upstream Status: Unknown
Origin: Gentoo
Description: Fixes building against FFmpeg version >= 2.0.0
}}}
would become:
{{{
Submitted By: Igor Živković <[email protected]>
Date: 2013-10-08
Initial Package Version: newer
Upstream Status: Unknown
Origin: Gentoo
Description: Fixes building against newer FFmpeg
}}}
Unfortunately, I got angry, and instead of making a logic discussion,
replied with a nonsense text.
Back to Alice, she told me: but it is a regression!
Then I sent her a link for a very similar previous discussion [3] about
gcc, where I found a solution [4], and apparently Bruce is not aware of.
Message of commit 13843 is [4]
"Create a standard for packages with problems introduced by gcc-4.9.0.
Modified: mdadm-3.3.1, gst-plugins-base-0.10.36, LAME-3.99.5 and
LibreOffice-4.2.5. Thanks Christopher G., from systemd branch, for
pointing that out, and Akhiezer and Armin K. for discussions."
"En passent", this clearly proves that Armin knew he was trying
something discussed before, but dissimulated it with a nice title, mixed
it with another fix, and it is known to everyone here tha patch in
ticket can be read only if clicked on it, from inside the ticket. He
also made a commit to blfs trunk and reverted (part of) it "Oh, it was
not intentional" to produce a diversion of attention to all those "en
garde". He got it!!! He succeeded!!! Congratulations to Armin for winnig
the battle and for the tactics!!!
What Alice prefers what we have for gcc: "For i686 systems, fix a
problem introduced by gcc-4.9.0."
This has about 100% of information, and as long is there, all newer
versions of gcc are also supposed to need the fix.
Incidentally, now, we have the following xml files with that gcc text:
multimedia/libdriv/gst-plugins-base.xml
multimedia/audioutils/lame.xml
postlfs/filesystems/mdadm.xml
general.ent
xsoft/office/libreoffice.xml
Alice told me that something should be done, or the book gets
unbalanced, inconsistent.
I agree with her: either we hide the information in the pages above,
necessary for the user to know if it is relevant, or fix the commit
14653, not to the previous state, but to what is now used for GCC:
- If building with --enable-freetype2 configure switch, use the
following
- <command>sed</command> to fix building against newer versions of
- <application>FreeType</application>:
+ If building with --enable-freetype2 configure switch, use the
following
+ <command>sed</command> to fix building against
+ <application>FreeType-2.5.1</application> or newer versions:
Without it, the book is much clearly wrong!
I would like to thank Alice and Alex for discussions that allowed me to
think in a more logical way, because thinking with anger is difficult
and leads one to hell.
I would like to rewrite that Bruce has the reason: I am tired, but not
less than when he stated it. I will try to do some work today, but speed
and correctness will suffer very much.
Finally, I wrote in the previous message:
>> From now on, I will keep doing my work in updates, as long as I am
>> not having headaches, but won't mind anymore if anybody changes
>> what I have done. It is not worth my mental health, having an
>> opinion and fight for it, so, I will try my best to not give
>> opinions about anything.
>> Also, it is not worth asking me about something someone thinks I
>> did wrong: my anticipated response, please know, is yes, you are
>> right and I am wrong, go ahead and fix it.
Therefore, most probably, I am wrong in everything I wrote in this post.
But who knows somebody with clear and logical think might tell what is
really correct.
--
[]s,
Fernando
[1] http://wiki.linuxfromscratch.org/blfs/attachment/ticket/5702/meh
[2] http://wiki.linuxfromscratch.org/blfs/changeset/14653
[3]
https://www.mail-archive.com/[email protected]/msg00738.html
and previous messages in the thread
[4] http://wiki.linuxfromscratch.org/blfs/changeset/13843
--
http://lists.linuxfromscratch.org/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
