2010-12-12

Acer again

At last I've bought it: a new Acer Extensa 5235... Once I said I won't buy an Acer again because it was unsupportive with GNU/Linux and too much sort of "Microsoft is the best, buy it with my computers", but it happened this one came without Microsoft Windows... Maybe (surely) not the best hardware I could get, but it is a lot more than I've got before, and cheaper than I planned at the beginning.

It came with "Linpus", that I wiped out after discovering a part of it (if not the whole system) was compiled in 2008, and that it was badly installed (in fact the computer started only selecting "start with the last working config" from grub menu), and that it wasted a lot of hd giving not too much... thus I've rejected the dual boot idea and reinstalled Ubuntu 10.10, and I am still "tuning" it (again:().

Having this powerful hardware, and a lot of room on the hd (going from 30 Gbyte shared among MS Windows and GNU/Linux to 250 Gbyte made me ask myself what I can do with all that free space...), I've installed many things I had not installed before, e.g. and notably FlightGear.

While installing software and software, I've realized I have badly partitioned the hd, and soon I'll have to change something; anyway this should not be an issue I hope.

Leaving some notes that could be useful:


  • To be able to control brightness from Ubuntu, I had to add acpi_osi="Linux" into GRUB_CMDLINE_LINUX_DEFAULT in /etc/default/grub (you have to escape the " and write indeed \"), and so in the same time I become aware of the fact that grub is changed: no more menu.lst, it's built on the run now...

  • If you install from "third parties" or, even worse, from files that are not .deb (like the Adobe Acrobat Reader which give an executable) and you are not able to edit your menus anymore, you could find interesting to know that the program for menu editing is called alacarte (from french à la carte, I suppose) and if you run it from the command line you'll be informed of what went bad; in my case, it happened that some directories and files in my ~/.config/menus folder had root as owner, instead of me... of course, a consequence of a badly written installer called with sudo; solved with sudo chown -R ME:ME .config/menus; check for symlinks before.

  • Webcam and microphone works without any effort, and wireless too...

  • ZynAddSubFx had problems and I've uninstalled it; though it is an interesting piece of software, I am now "devoted" to Csound and other tools, so I don't need it; however I think the knot that made it not work could be back when running other apps. I hope I am wrong.

  • Emacs started with a too high window (frame), I had to apply a fix in the .emacs; currently I've used a solution found in a blog, but I do not like it since it change the size after it was opened, so the user can see the change; however the solution tries to get the best size at "runtime"

  • Though this notebook was not sold with MS Windows, the keyboard has its logo on it... as usual, I've mapped it to Meta (mainly for Emacs usage); and of course the Menu key is bound to Compose

  • Euro and Dollar key near the "direction keys" do not work; xev does not "listen" to them; I will find a solution one day; but I find those keys mostly unuseful, it would be more interesting to map them to something more interesting than €$, even if their position suggests the binding can't be "critic"



I haven't copied data from the old good hd into this one yet, so I feel still al little bit "naked". Since the old laptop is out of order, I'll need to put its hd in a IDE-usb "bay" or something like that. I'll do.

2010-12-08

Traffic stats

I've just seen that there are stats about blog traffic! And I've just discovered a lot of traffic came from StackOverflow, a place I've almost discontinued to frequent — good place for practical solutions to practical problems, but absolutely deaf and blind for reasoning, "theoretical" analysis of hypothesis, exploring knowledge and so on (like any other Q&A site, no matter how much technical they call themselves)...

Anyway since it can produce traffic to this very "useful" blog, the piece of me allured by net popularity (!) is suggesting me not to forget the place... Now that I am thinking about it, I've left alone a war-speech about the output of a (intentionally) wrong piece of C code and distinction between "undefined" and "unspecified" behaviour in the standard paper for the language. It's enough for today, but in the weekend, if I'll own a brand new laptop, I'll poke the fire until someone can explain for real and competently why that code outputs what many people expect, as if it would be "unspecified behaviour", and what should realistically happen under the cover of the compiler in order to manifest the "undefined behaviour" — my position was that the specific code will output always what indeed it outputs, despite the "undefinition", and so, in the specific case, for the specific code, no "undefined behaviour" can manifest itself (and indeed even the unspecified behaviour would be hidden by the way the code was written, it was about passing a "problematic" argument to a printf with a format string that would take only one argument, details on SO, but I'll be back on the topic, don't be afraid).

Wo, two posts in a day, I am growing the statistics!

How many days in a month?

A coworker proposed to me a little challenge sometimes I think about. My first solution, though correct, was rejected since it used the ternary operator ?: and there's a way to succeed without.

Today is holiday here in Italy and I am using my spare time pushing the button of my almost-dead laptop just to hear the one-long-beep-two-short-beeps sequence (hoping not to hear it), then hold the button to turn it off and then stick hitting keys on the job laptop — where I am now.

After I've read a lot of pages from On the Edge - The Spectacular Rise and Fall of Commodore, almost finishing it, I decided to let my eyes be enlightened by the LCD of the job laptop, where there's nothing more interesting to do than surfing the net and using gcc (MinGW) in the unpleasant environment MS Windows XP provides.

Ok stop making the post longer than it could be...

The challenge was: using just an expression, get the number of days in a specified month (given as a number from 1 to 12). No-one told me about constraints for February so I assumed it has 28 days; moreover, initially there was no prohibition of the ternary operator...

My first solution was:

g = 31 - ((m < 9)? (m+1)%2 : m%2) - 2*(m==2);

which worked on all the system I've tested it (MS Windows, Sun Solaris on SPARC, and GNU/Linux on Intel) but obviously contains a possible bug dued to m==2 that does not need to be evaluated as 1 when true, though likely it does (As std C says, boolean expressions evaluate to 1 when true; unless we're dealing with non std compliant compiler, or if we translate the idea in other languages).

My second solution is:

g = "103012310321"[m-1] - 18 - (m&2);

This one is maybe tricky but is more robust (provided m is in [1,12] and we are on a system using ASCII), and can be easily modified in order to yield 29 for February. Moreover, what it makes me like it is that it uses something rarely many non-expert programmers are aware of about C (I mean using "array notation" for a literal string). The practice could be marked as bad, but it is great when codegolfing!

About codegolfing... If I am not wrong, that code in GolfRun would be "103012310321"m(=18-m2&- but currently I can't test it since my already-compiled (for GNU/Linux) work-in-progress GolfRun interpreter is on the Real Out-Of-Work Laptop, though I could checkout the SVN repository on this machine and try to compile it with MinGW on MS Windows — what a pity it uses Gtk+ and GMP extensively and I want not to go into the trouble of having them properly installed (ready for development usage) on this hateable system.

The third solution to the challenge that comes into my mind is about looking at the bits. The previous solution already does it... a bit! But the idea can be extended; for this purpose I've made the following table


XY ABCD
31 1F 1 11 11 1 0001
28 1C 1 11 00 2 0010
31 1F 1 11 11 3 0011
30 1E 1 11 10 4 0100
31 1F 1 11 11 5 0101
30 1E 1 11 10 6 0110
31 1F 1 11 11 7 0111
31 1F 1 11 11 8 1000
30 1E 1 11 10 9 1001
31 1F 1 11 11 10 1010
30 1E 1 11 10 11 1011
31 1F 1 11 11 12 1100


We have four bits in input and we need to "produce" just 4 "patterns" (two bits) in output. The boolean table old school (and semplification thereafter) will give us a solution (paper and pen here for a while...) (^ stands for an overbar on the following letter, i.e. for the logical not):


^X = A + B + ^C + D
^Y = ^A^D(B + C) + AD^B



That gave me the code:


g = 28 |
((((((((~m)>>2)&((~m)<<1)&((~m)>>1))^1) &
(((~m)^2)&((~m)>>3))) |
((m>>3)&((~m)>>2)&m&1))^3)&3);


I am rather dissatisfied with this: surely it can be optimised someway, e.g. by doing simple observations like: no need to handle higher bit in logic since it needs just an exception (when m is 2); try to work in "positive logic"; mix ideas from the previous solution with the "gate logic" stuff... Anyway, for now it is enough, I have to go and do better things: I have a button I must push.

2010-12-04

Bye bye TM250

Maybe 1 post per month is a very low activity for a personal blog, should I be more chatty? Ok, let's say so: car broken, a lot of money to make it work again but this is not hard to handle. The thing that is hard to handle is to accept that my laptop is old, and it started to stop working. Well, not exactly. It's hard to accept the fact that it indeed works still well, except for a small detail that makes it unworking most of the time. If the diagnosis is right, it is about the BIOS EPROM... CPU is ok. Hard disk is ok (good! no data loss for now), LCD works, fan spins, some keys is a little bit deaf but not too much, and I have an external usb keyboard... RAM is ok (I went from 256M to 1G, remember?), gfx card is ok... everything's ok, but that little piece, which decided to fool me (if the diagnosis is right, again).

It happened one day when instead of seeing the Acer logo screen, and then the Grub boot loader, a long annoying beep followed by two short beeps greeted me... What's up? A lot of BIOS says graphics card problems this way. But a deeper seek made me informed about the fact that 1L-2S is (should be!) a problem in the flash of the BIOS... Likely.

Somewhere I've found a solution: reflashing the ROM, twice or three times to be "safer", and everything should go... Not so for me (or this was not the problem...?) When luckly the PC started, after 20 or so attempt, I tried the procedure. It seemed ok for a while: turned off, restarted (after minutes), and it was ok. Hurrah... but then I waited longer and the problem showed itself again.

This time it took longer before a turn-on attempt was successful. Reflashed (using provided software) the Phoenix BIOS 1.15, ten times or so, but likely this won't solve... Now I am using the laptop, since I only rebooted between MS Windows and Ubuntu, never turned it off (yet...). And it's all ok. I know tomorrow it won't boot, I will try and try, and at some point I'll give up. And I will decide to buy a brand new laptop... In italian there's the slang local verb "rosicare"; it should exist a synonym I can then translate into english, but in this moment it comes not into my mind... it is the feeling you can taste e.g. when you are so close to a goal and then, suddenly, a small stupid detail goes wrong and you lose the goal.

My pre-euro-era laptop still works... except for that stupid detail that if fixed would make this piece of hardware last for many years still ... one year would be enough for me!

Useful post, isn't it?

2010-11-13

Drop a drop

Finally my badly over-tweaked "old" Mandriva's got wiped out by a fresh brand new Ubuntu 10.10. With the advent of a bit more of memory (from 256M to 1G!) I've decided to try again a full featured modern desktop and accept all the comforts of the defaults, though a little bit too much windowsish and even though accepting defaults is usually a bad idea. There will be time to tweak this system too and make it usable just by me...

In the meantime I was disappointed by a really annoying bug I've noticed while watching a flash video on YT. At first the guilty, to me, was the flash player... But soon I've discovered it was not. Submitted a bug report for "choppy audio", ... but recently I've realized it was not about it. The bug is bigger: the whole system is stuck "occasionally".

Today, I've timed the choppiness and surprise, it's very regular: one "stuck" every ten seconds. I've searched the net simply with "ubuntu stuck every ten seconds" and found this link very useful. Then I tried the nomodeset and woah now it is ok. On an Ubuntu wiki I've found how to disable permanently it by adding a file in /etc/modprobe.d containing simply an option for the i915 driver: options i915 modeset=0. Indeed I've not restarted the system yet... however, if the boot option worked, the permanent solution should too.

This is however still an annoying bug. I don't know if KMS is really good, but its problem with Intel chipset should be fixed.

Another thing I did on Mandriva was to kill the tslash from the key t, and put in its place the þ. I don't know way in the "latin" layout (type) for danish, norwegian (etc) the þ is where it should be, while otherwise it is bound graphically (instead of phonetically) to the p key.

After reading here and there about how to use xmodmap, and the .Xmodmap file in the home, I've decided to fix it my way, as I did in Mandriva: editing the file /usr/share/X11/xkb/symbols/latin (maybe on Mandriva was located elsewhere, I can't remember). Did it and now that I've mapped (this time using the provided configuration tool) the Menu key to the Multi key and the unuseful Windows key to Meta key, I feel better.

The next step will be to fix a bit Nautilus: I don't like the Windows-like behaviour when you double-click on files it "does not know". I was used to be able to open all files (known or not) with a text editor (Emacs usually) or an hex editor (ghex2), but it seems that I have to pass to the "windows-steps" to do it: if a filetype is unknown, pick a program from a list, remember the association (no, never!) and of course if the file is "known", the context menu will show the item to open it with the program that "understands" it. It can be ok, but I want also always the items to open it with an editor (text and hex). How do I do it?

From the packages point of view, I was able to install everything I needed; in particular I was worried about interpreters/compilers I had installed on Mandriva. Luckly I was able to install maybe all what I had, from C (of course!) to Algol 68 Genie, with apt-get or compiling it (MMIX, MMIXX and Algol 68G only, currently, while the rest was all found in .debs in the repo or on the net). Rebol being proprietary is still not installed, but I will; and also fasm (flat assembler) is missing (but of course there's nasm).

2010-10-09

My own thoughts about software development methodologies

work in progress draft
and it always will be


I wish to scatter some thoughts about the subject. There's no a particular order, deep analysis or claim for correctness.


  • Defining the project and its requirements


    • Do not let the developers define the requirements! Requirements come first, then it comes developing code (with feedback, if needed/required).


  • Documenting APIs (if appropriate)


    • Be brief and effective, do not add unuseful redundancy; it must be techinical, even when it is a "high level" API

    • Let the dirty work be done automagically: force programmers to document into the code, using a special, fixed syntax (well-known or created ad hoc), but no literate programming — maybe with exceptions to be defined

    • APIs' docs must not expose internals; if you need internals, put them in a separate document

    • References and links, if appropriate; no copy-pasting

    • If there's a protocol to access the API, describe it in a formal way (if it is a standard, just reference the standard document describing it, avoid giving your own incomplete simplistic description); if needed, train whoever has to understand the formal description, rather than sticking to a informal loosy description — which however is suitable as introduction and to give a general picture, if appropriate

    • Do not use a format which is not mechanically parseable and that it is hard to be generated automatically. The best choice is still plain text with an agreed encoding (e.g. UTF-8) and line ending convention (e.g. LF, the "unix" way to terminate lines); however, for documents generated by source code, ASCII is likely the safer choice. TeX-based or HTML-based conventions for formatting are good ideas.

    • Deploy the doc to the "rest of the world" in a read-only format, like PDF (which is likely also the best choice) — this implies the ability to generate the PDF from the "private" doc.

    • Avoid translations: if your doc must be read "internationally", do it directly in english; in Europe, if it would be widespread, esperanto could be a fine choice... but alas it is still english that it is better to use! If people working in the project can't write/read "technical english" (to be distinguished from "literary english" and real-world english), teach it to them!


  • Writing the code...


    • Conventions! Train programmers to use an agreed set of conventions (tabs, spaces, indenting, symbols' and variables' names, comments' style...)

    • Comment, but not too much; avoid "line by line" comments, prefer "logical block" comments

    • This is not always true but... more data structures, less code!

    • Break functionalities into as small as possible functions/objects (depending on programming paradigm in use)

    • Think functions as reusable bricks, leave room for enhancements without breaking existing working code (this is easier in OO programming, nonetheless it is not impossible in non-OO programming)

    • Refactor your code; possibly before it becomes a mess; even when this happens, refactor it anyway, it is not a waste of time/money (at least not in the long run), it is needed to keep your code manageable and...

    • Refactor before your code becomes unmanageable, and refactor/redesign if your code is already hardly manageable; this sometimes could be done in small steps while developing other features

    • Use well-known (in the community of your language) libraries, rather than your own implementation (if there are no licensing issues); but of course if you can't find what you need, take your time to implement it rather than sticking to other possible but poor solutions

    • Don't waste time for optimizations, if your bottle-necks are elsewhere! (Profile your code to see where it is slow, and optimize there!)

    • If your language misses a hashtable and you need to choose actions according to a "pattern" given as input, ... find a hashtable implementation/library rather than using a long list of if-then-elseif or a C-like switch. Match this suggestion with the optimization topic if appropriate; if not, consider that the hash-approach produces clea(n/r)er code, as "more data structures, less code" does! (In fact the specific hashtable example is about following the suggestion "more data structures, less code", avoiding linear search.)

    • The previous reasoning on the hashtable is applicable to other features too, with the appropriate changes

    • Avoid copy-pasting; sometimes it is not so bad, but if you do it or need it too often, it means your code is badly designed/implemented... time for refactoring!

    • Test functionalities apart; e.g. write "toy code" and check its output given a specific input

    • Do not use so called Hungarian Notation; if to you it's an issue, use a good IDE to help you!

    • Use meaningful names for symbols/variables; but iterations in language having no special syntax to iterate over a "collection" (or a slice of it) can (should!) use short common names for indexes (like i j k n m).


2010-09-12

The "basic" thing

For a project with friends of mine I was requested to write a C# class. Ok, it is a good opportunity to learn in practice some bits of that language, I thought. But at the end what happened was that large part of the C# code has to be generated automatically by a VBA script (I wanted to do it in Perl, but since the code must be generated from MS Excel data and formulae, VBA was suggested as a easier and more microsoft-friendly solution), and so currently I've written more VBA code than C# code!

At the very beginning I was almost disgusted, but now I've realized that after all it was funny. It remembered me ancient days of "basic" structured programming, and having limitations that push you to find "workarounds" is funny too — though scaring if you have milestones!

I would like to give more time to C# enhancing the google video downloader, which is still a horrorful immature crap (though working for what it's worth); nonetheless now I can add VBA to the list of languages I've frequented someway... just another brick in the wall.

2010-07-20

Syntax and efficiency in C

Time for a brand new post.

In my last job interview I was asked to write a function to copy a string. Thinking not about compactness, I wrote on the fly the following code


void strcpy(const char *s, char *d)
{
while( *s != '\0' ) {
*d = *s; s++; d++;
}
*d = '\0';
}


That of course works, but it is longer than the need (and by the way, it is not compliant with what standard says about real strcpy, but this is not interesting since making a standard-compliant function was not the point of the question).

As noticed by the person making the interview, such a copy can be written in one line. When I've finished the interview and went back with the mind to what I've said and what I could have said, I realized that I could have defended my on-the-fly implementation: it is clearer (isn't it?), it has the same efficiency of the "compact" version, being this one possible just because of the great flexibility of the C syntax... which is, however, a boomerang if abused.

Second thought. The "long" code is clearer. Maybe (usually, not using C compactness ability makes code clearer; but likely in this particular case the one-line version is even clearer). But, my doubt is now: is it really the same with respect to efficiency?

Let us first see how to "convert" the long version to the short one. First, post-increments can be done "in place".


*d++ = *s++;


Then, assignment can be put into the while:


while( (*d++ = *s++) != '\0' ) ;


And here's the trap. We don't need the final "terminator-missing fix". Apparently, the short version looks also more efficient. But if we unroll the loops, we have the same amount of assignments, of course. Moreover, the short version assign the terminator reading it from the source; this gives two memory access (one for reading from s, one for writing into d).

On the other hand, the long version, does not need the extra reading access to memory and can be optimized on many processor with a single instruction writing a byte constant (read at the instruction fetch time).

So, it seems that is is the opposite: long version is (slightly) more efficient!

Of course modern compilers' optimization can cancel this apparent advantage. But let's look what gcc does with the basic "implicit" optimization level.

The "short version" (S) x86 code (with added comments):


mstrcpy:
pushl %ebp # prolog
movl %esp, %ebp
.L2:
movl 8(%ebp), %eax # get s in eax M
movzbl (%eax), %edx # *s in edx (dl) M
movl 12(%ebp), %eax # get d in eax M
movb %dl, (%eax) # *d = *s (dl) M
movl 12(%ebp), %eax # M
movzbl (%eax), %eax # take char*d in eax M
testb %al, %al # I
setne %al # is not 0? in the same al I
addl $1, 12(%ebp) # these are the post M
addl $1, 8(%ebp) # increments of d and s M
testb %al, %al # check al I
jne .L2 # loop if not 0 ('\0') JT
popl %ebp # epilog
ret


The "long version" (L) x86 code (with added comments)


mstrcpy2:
pushl %ebp # prolog
movl %esp, %ebp
jmp .L5 # uncond. jump
.L6:
movl 8(%ebp), %eax # get s in eax M
movzbl (%eax), %edx # *s in edx (dl) M
movl 12(%ebp), %eax # get d in eax M
movb %dl, (%eax) # *d = *s M
addl $1, 12(%ebp) # d++ M
addl $1, 8(%ebp) # s++ M
.L5:
movl 8(%ebp), %eax # get s M
movzbl (%eax), %eax # *s in eax (al) M
testb %al, %al # I
jne .L6 # exec body if not 0 JT
movl 12(%ebp), %eax # get d M
movb $0, (%eax) # *d = '\0' M
popl %ebp # epilog
ret


The M marks instructions that do memory access; let's count only the in-loop instructions first. S has 8 memory access instrs in loops; L has 8 memory access instrs in loop but S has 3 "register" instrs in loop, while L has none. So for S we have 8+3 in loop, for L only 8.

In S, when the '\0' is reached, the whole 8+3 instruction are performed; then the jump is not taken (JT stands for Jump Taken most of the times). In L, when the '\0' is found, (2+2)+1 instructions are executed (2 M after the .L5 label, and 2 M after the conditional jump). In fact S performs all the "actions" of the C code (post-increments inclusive) in the while; but L does only the test on *s and the final termination of d. (This means also that if we access s and d in the S version, they point one char beyond '\0', while in the L version we do not increment them, since we exit the loop, so s and d will point to the '\0').

On modern processors it is hard to say which is really more efficient. Likely, the difference in performance is negligible. Moreover, optimizations done by the compilers may level down the difference, if any.

It is however worth noting that L version does not produce a so longer code (considering also loop unrolling), and it could be even more efficient than S version!

2010-06-17

Back to hardware

When I was very little I was used to copy a lot of circuits from electronics magazines my father bought, and also from his own diagrams. To me they meant nothing at all, but I found them beautiful and I sensed that there was a meaning in all those symbols. Growing, strangely my interests shifted more towards software, even after I studied those symbols and assigned their proper meanings. At some point in time, I can't remember exactly where, I thought it would have been cool to be able to create a computer; I focused this idea when I studied more deeply digital electronics, but indeed I never planned to realize it for real.

Now it is almost time to get back the idea... It is renewed as something a little bit different: since to create the hardware for real is pragmatically hard (maybe FPGA could do, but...), what if I create a software simulation? The simulation "models" the digital components that "are" the microprocessor, properly interconnected.

Excited by this approach, that seemed possible and not so hard, I've created a new task for Rosetta Code, Four bit adder. The C "model" seemed to me to work greatly, I thought "It can be done", but the task is just a very short step towards the final aim, proved to be not so well written, and the code showed some limits this morning, when I've taken a look at how I could implement a latch using basicall the very same "ideas" already used in the four bit adder.

A little bit demoralized, I began searching the net for similar stuffs... And of course I've found very interesting things, and more "realistic" and done by people more gifted for hardware — I used SPICE once upon a time, but indeed never ever gone into hardware projecting for real, universitary laboratory apart, where however we did not too much complex circuits, in fact we have not done any ambitious project at all; and I know of the existance of languages like VHDL and Verilog, but never used them (I started to take a look at them since today).

I am swinging between two position. In one position, I want just a "simulator" realized in a known language like C (even though OO languages in this case could make things easier). In the other position, I am taking into account the hypothesis of using VHDL for real... and this second position started to exist when I've installed FreeHDL, which has a VHDL to C++ "translator"!!

So even my idea of writing the simulator in code is not so new... Well, the original part could be that I would like to realize an emulator for the processor, using as core the finished simulation. In practice, I would like to run something like myemul ROM where ROM is a file that provides the ROM, and myemul does not realize the emulation as Bochs or QEMU (which have performance needs), but realizes it at a low-level, through logic gates emulation...

Ok, too much for now. I've installed FreeHDL, gEDA, GtkWave and also ngspice. I think I have everything I can have for free (and mostly free, too). But I still does not know VHDL and Verilog, and I have not the slightest idea from where to begin. Luckly, the net is full of informations and experiments, like The Harp Project.

Note: discussing several arguments on Rosetta Code about the task (and a replacement for its description, since the current does not satisfy me) I've discovered that there could be a certain interest in having Brainfuck code running in gate-logic; maybe it would be not a bad idea to start projecting the processor so that it can "host" BF easily.

2010-06-10

Q/A sites

Question/Answer sites are always the same. I mean, there they happen always the same kind of things that can be summarized with few words: biased subjectivity, misunderstandings, personal battles using the voting system. I am not saying I don't fall in those traps too, except for the last one and the first one(!): I do not use the voting or flagging system to fight my personal battles or to "destroy" an answer that does not convince me, but which I recognize as having its own "rightness". Instead, I use often the possibility of commenting (our freedom of speech).

My english is far from being perfect, perfectly unambiguous and grammatically correct, maybe sometimes the hurry makes it even worse...; nonetheless it seems to me it is not so bad. But it seems to me sometimes I get misunderstood because of it.

Recently I am losing a little bit of my time on StackOverflow. Appearing as a site for technical questions, it seems free from all those absolute crap often existing in many low level Q/A sites (I mean question like: where I can get The Program registration key, how can I cancel a file, how can I program a word-processor like Commercial-Wordprocessor in few lines of code, ...), and in fact it is.

But as said, it is run by people, who are defective by design and so the typical aspects of social interactions can arise, altogether with all kind of possible misunderstandings and arbitrariness.

I have a good example where of course I am the protagonist.

The StackOverflow (SO) question is here. My original answer was downvoted by someone (later the downvoter declared himself and it was the same asker), and this was disappointing to me since another answer, containing basically the same kind of informations/proposals, was voted up (I don't know by who).

I loose time, ok, my choice, but I am not prepared to loose time for an answer that is not appreciated and moreover I can't understand why, since compared to others (voted up) it appears of the same quality. So I wrote a comment, knowing no the voter nor understanding the reason.

The comment was:

can I vote myself up? I find this -1 really stupid, expecially since basically I am saying the same thing of an answer that's got +1 and say no word at all about refactorying the if with a subroutines. I've already said that: no matter the techinicalities, Q/A sites are frequented by the same kind of voters.


and I started writing it before the asker comment (As stated, CAN'T CHANGE RULE) appeared. The long "dialog" that follows in the comments came from two facts: my distraction, and their distraction. As usual I am not interested in pointless debates, I just try to understand.

My recognized distraction was that he stated already he could not modify Rule class, while unluckly I focused my "defense" on the fact that he did not, instead of focusing the attention on the fact that, no matter if he specified it or no, my answer contained already a possible solution that fit the requirement, exactly as another answer. Indeed, better than another answer, the one that was upvoted anyway!!

The original asker felt uncomfortable with my comments and decided to remove the downvote (since it was causing "issues"), later pointing out however that the requirement my answer did not talk about was in the original post. Unluckly I noticed I proposed an answer also for that requirement, but not focused it in time, and another thing that happens on Q/A sites is the lost of attention where things get complicated or too long to read.

As the following Original Poster comment shows, he focused on the "first" part of my answer, which talked about changing Rule class (exactly how another answer did, but was upvoted... again)

My downvote is because that your answer ignores my question. Yes, refactoring the Rule class to use an int would be useful, and in fact, I could keep the "else if(flag == "EXCLUDE") { included = false" given your example. So it's a very good answer to a completely different question


The fact is that my original answer contained already at the beginning the following excerpt:

Now if you want the whole [CODE] as a subroutine, you can make it return boolean value and check for it. The caller "decides" if break/return or not. Of course the caller must pass i also (if looping over i for get_op changed as described before).


This shows that indeed I've not totally ignored the requirement. My if looping over i for get_op changed a described before means exactly that: if (i.e. you are not required to) you change get_op as described, you have to pass i. If you do or not, no matter: you can turn the [CODE] into a subroutine returning true or false, and according to this returned value, the caller decides if to return directly or continue looping.

Moreover the OPer edited the question and added a code that basically do what I suggested, wanting as arguments get_opN and get_niN instead of N to allow the subroutine to pick the right get_op(N) and get_ni(N), but apart this detail (due to the requirement) the code is what one programmer can take from the suggestion I quoted before.

While another downvoter chimed in my answer, another up vote was given to the answer that contained same suggestion of mine (and no citation at all to the subroutine possibility); the author of this answer "defended" me some way, thanks. But of course my critics are not against him, but the illogical incoherence of the voters...

The brand new downvoter wanted also to explain its -1. And someone put +1 to its comment.

@Shin: I think the OP's downvote was fully justified. You wrote an answer entirely based on an assumption that the question clearly specifies is invalid. That means your answer doesn't solve the problem.


This confirms that jalf (the nick of the previous commentator) coherently should have downvoted another answer too (the one similar to the mine), but he did not. He downvoted mine instead, because of comments. Talking about what it can be just or not, it is a subjective thing. But also an imperfect "justice" needs to appear coherent: if A-by-X is wrong, it must be wrong also A-by-Y. An answer can't deserve a downvote where another answer, saying the same thing, deserve an upvote. A coherent "just" person can't vote down an answer A-by-X and let untouched A-by-Y (and A-by-Z and so on...). If he does, the only explanation brings us to the fight personal battles using the voting system, which is unfair. In this case, my "complainings" likely was what have driven jalf to the downvote. He did not evaluate the answer deeply, otherwise he wouldn't have used fully and entirely, and he would have noticed that I talk about subroutine with an if that allows for the other explicitly missing case. At least another answer doesn't solve the problem (here of course it should have been "does not try to solve the problem", or I should see that jalf has a huge amount of downvotes!), but is ignored, meaning this is not the real point.

Unluckly when I read that biased comment I have not realized what's wrong in all the "defense line" (i.e. the fact that indeed I treated also the case matching the requirement)... and when I did, it was too late: attention focus was lost and likely jalf won't realize and recognize his error, and this is normally what makes me put an user on my personal black list (in the sense I have to watch out when they are around, to avoid wasting time and get also downvoted unjustly).

Talking about incoherence, it could seem incoherent the fact that I am focusing on jalf and not the asker. It is not because he retired the downvote (I did not pretend him to do so, but to explain, at least why he did not downvote similar answer); it is because he interfered in the dialog like if he was the Knight of Justice (Restoring the Justice, i.e. the just downvote by the asker), unaware of his incoherence and moreover without reading carefully what was already written. People like that rarely come back and re-ponder, it would be wasting too much time on a single question, or even worse: on comments about an answer to a question.

I am not saying jalf is a bad guy to me of course. I am saying he bahaved the way I don't like people to behave on Q/A sites. Commenting is fine. Voting up or down should be more pondered on the current answer, not biased, and coherent. Since these people could make my "wasted time" less pleasant, I mark them for my black list.

The irony is that the asker mixing up answers or by himself produced a code that is already an answer, and that is contained, as idea, already in my criticized and downvoted answer (and not in the one similar to mine and upvoted)... so that my last vote mixed everything and proposed a solution, which fits fully the requirement. But jalf has not wasted time checking for updates, and no other readers have (currently) upvoted me, except who did it at the beginning (and gave the upvote purposely because of the part of the answer that does not meet the requirement, which is a poor bad requirement after all!)... while as said the other answer, talking no about refactoring in subroutine at all, has 2 votes!