1 |
On Tue, Feb 18, 2014 at 11:06 AM, Andrew Savchenko <bircoph@×××××.com> wrote: |
2 |
> On Mon, 17 Feb 2014 23:30:42 -0600 Canek Peláez Valdés wrote: |
3 |
>> On Mon, Feb 17, 2014 at 8:05 PM, Gevisz <gevisz@×××××.com> wrote: |
4 |
>> [ snip ] |
5 |
>> > How can you be sure if something is "large enough" if, as you say below, |
6 |
>> > you do not care about probabilities? |
7 |
>> |
8 |
>> By writing correct code? |
9 |
> |
10 |
> Real world code without mistakes and larger than "Hello, world!" |
11 |
> exercises is not possible. Large systems must have error suppression |
12 |
> and correction techniques, modular and replaceable design is one of |
13 |
> them, KISS is another one. Systemd has none known to me. |
14 |
|
15 |
It is modular. It is simple under the (larger) scope it tries to cover. |
16 |
|
17 |
It will have bugs (like *any* other non-trival program, as you said), |
18 |
obviously; but in time those bugs will be fixed and everything will be |
19 |
fine. |
20 |
|
21 |
>> >> I don't care about probabilities; |
22 |
>> > |
23 |
>> > If you do not care (= do not now anything) about probabilities |
24 |
>> > (and mathematics, in general), you just unable to understand |
25 |
>> > that debugging a program with 200K lines of code take |
26 |
>> > |
27 |
>> > 200000!/(10000!)^20 |
28 |
>> > |
29 |
>> > more time than debugging of 20 different programs with 10K lines of |
30 |
>> > code. You can try to calculate that number yourself but I quite sure |
31 |
>> > that if the latter can take, say, 20 days, the former can take |
32 |
>> > millions of years. |
33 |
>> > |
34 |
>> > It is all the probability! Or, to be more precise, combinatorics. |
35 |
>> |
36 |
>> My PhD thesis (which I will defend in a few weeks) is in computer |
37 |
>> science, specifically computational geometry and combinatorics. |
38 |
> |
39 |
> You're not the one here on the list with PhD (either defended or |
40 |
> near its end). And argument "Listen to me! I'm PhD here!" looks |
41 |
> miserable. Please stop this. |
42 |
|
43 |
And you please stop twisting my words. I mentioned my background only |
44 |
because someone was trying to teach me about combinatorics (which has |
45 |
nothing to do with this, BTW). |
46 |
|
47 |
It was not to give any weight to any other argument. |
48 |
|
49 |
>> >> I care about facts: FACT, I've been using systemd since 2010, |
50 |
>> >> in several machines, and I haven't had a single segfault. |
51 |
>> > |
52 |
>> > Have you ever tried forex? If yes, you should have been warned |
53 |
>> > that "no past performance guarantee the future one." |
54 |
>> |
55 |
>> I never said that. I trust the quality of the code, measured by my own |
56 |
>> judgment and bug reports, etc. Not past performance. |
57 |
>> |
58 |
>> And even if a bug goes by, then what? The world will end? |
59 |
> |
60 |
> This depends on what bug at what component occurred. Just imagine |
61 |
> pid 1 segfault on medical life support equipment. With systemd going |
62 |
> into embedded this is not just pure speculation, though, of course |
63 |
> medical stuff should have extra safeguards. But any FT or at |
64 |
> least HA setup is a combination of multiple layers. I do not want to |
65 |
> allow badly broken core component on mine setups even if its faults |
66 |
> may be compensated by other means. |
67 |
|
68 |
There are stable releases and testing releases; you put in |
69 |
live-dependent code the most rock solid software you have. The one |
70 |
that has been thoroughly tested and used. |
71 |
|
72 |
Really, "small" and "modular" don't guarantee anything either; anyhow |
73 |
you need to test it and use it before putting in live-dependent |
74 |
systems. |
75 |
|
76 |
> Yet again, I respect ones right to use whatever one wants, but I ask |
77 |
> to respect mine as well. That's why I propose a separate systemd |
78 |
> profile for those willing to use it. |
79 |
|
80 |
Then write. Just be aware that to write a systemd profile, you need to |
81 |
use systemd. |
82 |
|
83 |
>> >> Sorry, but it's you who doesn't know the matter at hand: kdbus was |
84 |
>> >> (and is) written by Greg Kroah-Hartman, Linus' right hand, and who |
85 |
>> >> works for the Linux Foundation. |
86 |
>> > |
87 |
>> > Lol, he seems to start to use the arguments like "You even do not know |
88 |
>> > my elder brother/acquaintance from the street nearby who can easily |
89 |
>> > hit you down!" |
90 |
>> |
91 |
>> If you don't think Greg's words have any weight in a Linux-related |
92 |
>> technical discussion, then I'm afraid we will need to agree to |
93 |
>> disagree on any technical subject. |
94 |
> |
95 |
> You know, common sense should always override person's prestige. |
96 |
> History knows many examples. Sir Isaac Newton enforced corpuscular |
97 |
> point of view on the light's nature. And while he was genius in other |
98 |
> physical aspects, he was mistaken here. Albert Einstein was rejective |
99 |
> to probabilistic nature of quantum mechanics and even proposed an |
100 |
> entangled particles paradox as an example of its "flawed" nature. |
101 |
> Though as we know these days such systems exist and are quite well |
102 |
> used in numerous experiments. My point is simple: do not blindly |
103 |
> adhere to someone's words, even if this person has high authority. |
104 |
> Common sense must prevail. Period. |
105 |
|
106 |
I don't blindly adhere to *anyone* points. There are arguments that |
107 |
Greg, Lennart and Kay had made which I don't agree with. |
108 |
|
109 |
But in the big picture I'm with those guys. And so it seems the |
110 |
majority of the Linux world; you say it's because monetary reasons. I |
111 |
said it is because of the technical advantages of systemd. |
112 |
|
113 |
Regards. |
114 |
-- |
115 |
Canek Peláez Valdés |
116 |
Posgrado en Ciencia e Ingeniería de la Computación |
117 |
Universidad Nacional Autónoma de México |