Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

alpenwasser

Senior Moderator
  • Content Count

    4,123
  • Joined

  • Last visited

Awards


About alpenwasser

  • Title
    2CPU Enthusiast

Contact Methods

Profile Information

  • Gender
    Male
  • Location
    Switzerland
  • Interests
    all things mechanical, PC's, Unixes, hiking, camping, history, paleontology, biology, movies
  • Biography
    born (1985) -> school -> army (Lt, Infantry) -> work (logistics, web dev) -> electrical engineering
  • Occupation
    Electrical Engineer (DSP, Embedded Systems)

System

  • CPU
    2 × Intel Xeon X5680 (6C/12T per CPU), 3.33 GHz
  • Motherboard
    EVGA SR-2
  • RAM
    24 GB Corsair Dominators 1866 MHz
  • GPU
    GTX Titan & GTX 780
  • Case
    Caselabs SMH10
  • Storage
    40-ish TB server w/ ZFS
  • PSU
    Enermax Platimax 1200 W
  • Display(s)
    Dell 24" 1920x1200
  • Cooling
    2 × XSPC Raystorm Copper, MIPS block for EVGA SR-2, 2 × Aquacomputer Titan, 2 × HWLabs SR-1 560 mm, 2 × Alphacool Dominator RAM blocks, 2 × D5, 1 × Alphacool XT45 480 mm, Aquacomputer Aqualis
  • Keyboard
    QPAD MK85 w/ Cherry Blues
  • Mouse
    Wacom Intuos
  • Sound
    Yes
  • Operating System
    Arch Linux

Recent Profile Visitors

18,248 profile views
  1. alpenwasser

    Why the Linux CoC is Bad

    The dude seems like he has some serious anger issues.
  2. alpenwasser

    Why the Linux CoC is Bad

    I see, thanks for that link. I think the trouble with meritocracy here is what people understand it to mean, and how much importance they place on it. Most people I have talked with about the topic have focused very heavily on technical skills, neglecting or even entirely omitting social skills. That's also why I have been somewhat conditioned to it being a red flag for me, and I'll usually try to steer clear of environments which are obsessed with it. If somebody defines merit so narrowly, and then places it above anything and all else, I'm going to have to disagree with that. As you yourself said: People skills matter a lot. If we define merit more broadly, to encompass many more aspects of the human experience, sure, it might become more useful. But even then ... who decides what merit is, and how? Certainly, there are many metrics you can measure, but how to decide which metrics matter, and how much? And yes, you can notice if somebody has good people skills (or rather: somebody can convince you that they have), but it's not like you can just grade that and give them a 8/10 and that's a truly objectively correct measurement. And even in cases where clear metrics exist, how often and how rigorously are they actually applied by the decision-makers? You might be able to measure contribution frequency, compile times and execution speed, but when it comes to things like legibility, commenting quality and beauty of code, things are suddenly more subjective. But are those criteria irrelevant? At least to me, they're not. Your mileage may admittedly vary. So as soon as you have things like that being taken into account (and I think they should be -- people may disagree with me on that, fair enough), you suddenly no longer have a truly objective measurement of a person's merit (and I think objectivity is a goal of a meritocracy, at least for the people with whom I have so far talked about this in my life). People contributing to different parts of a project under different supervisors will not be judged equally because they're not judged by the same people. I mean, just look at this forum: Sure, us moderators try to make the user experience as uniform as possible by talking to each other about problem cases and warnings before taking action, but even in a team which is relatively small compared to the Linux project, we do not always succeed, despite best intentions. And that's from people who are making an active effort -- people in other projects might not even bother with that. And just like with the CoC itself, there's plenty of potential for abuse if the person above you in the food chain is an asshole who doesn't like you. Even if their criticisms are invalid and can be disproved, should it really be upon you to invest the time and energy to do that if you're donating your spare time to a FOSS project? Personally, I just wouldn't bother in such cases and stop contributing. I have enough interesting stuff to do in life where I don't have to deal with abrasive nincompoops. So I suppose one of my main gripes with the term (not the concept, just the word itself) "meritocracy" is that it suggests a standard of objectivity which usually just does not exist. It's not a magical silver bullet which automatically leads to great results, but to many people (at least ones I've talked to), the word sort of implies precisely that (whether or not those people are correct in their assumptions is a different question of course). Yes, what we end up with in practice is sort of a meritocracy as well, as already said, but the term itself has very different connotations for many people. And as said: The CoC just by itself doesn't really ellicit much emotion from me. There isn't really anything in there which I'd consider spectacular news. If she uses it as a tool of power, then that's more of a problem with her and the people listening to her, less with the bare text, I think. Most CoCs could be used for that, probably including our own. Sorry for wall of text.
  3. alpenwasser

    Why the Linux CoC is Bad

    I haven't followed her closely enough to be familiar with her views and actions in detail, so I don't know what she personally advocates for. But I couldn't actually find anything like this in the actual new CoC itself. Am I reading the right document? https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=8a104f8b5867c682d994ffa7a74093c54469c11f I've read that document multiple times now, and honestly, I don't think it's very spectacular, or anything worth getting upset about. Unless I'm reading the wrong thing. IDK. As for your actual question: If my boss said to me "[Person X] is better because they are [part of minority Y].", then I would obviously consider that preposterous. But I don't see anything like that in the CoC. If the author of the CoC advocates for that, that's not really a fault with the CoC itself I think. But maybe I haven't found the right link or right document or am reading the wrong text, or have overseen a sentence or paragraph. Weirder things have been known to happen. It's easily possible that it's going to be badly enforced, but that's possible with any CoC -- even ours. It just depends on the mods, as @Sauron says. It's not a problem with the text, it's a problem with its application. In any case, most of the time, I try to stay way from FOSS drama -- too many fragile egos to deal with. My life's too short for that.
  4. alpenwasser

    Why the Linux CoC is Bad

    Feels great, to be honest. I even get paid decent money for it. More seriously though: My point is less that merit in and of itself is useless, but that most of the time when people go on about "meritocracy", they focus almost exclusively on the technological side of things, when in practice, whether or not you hold merit to a specific project is determined by many more factors. As said: An engineer with crappy people skills, no matter how brilliant on a technological level, would not fare well in the environment in which I work. They would be useless; i.e. without merit. Sure, what we end up with is a meritocracy of sorts as well. But not the kind of meritocracy of which your average internet dweller will think when they read the term, most of the time. I'm not saying my tech skills are irrelevant (otherwise I could be replaced by a preschooler, after all). But they're by far not the only thing which is required for me to be good at my job. So applying the term to my situation isn't very useful, because people are going to get the wrong idea. We need engineers with good people skills on top of good technological expertise where I work, otherwise we're simply going to fail at our mission.
  5. alpenwasser

    Why the Linux CoC is Bad

    As a side note on meritocracy: It sounds great on paper, but I've found it to be a rather useless concept in practice. Primarily because what "merit" is exactly in any given context is not as straightforward to determine a lot of the time and open to interpretation and ambiguity (at least in my experience). Example: I work in a relatively small team of engineers at my company, but we're involved in some EU and German research projects where you work with lots of people from other companies and universities. Beyond a certain baseline of technical competency and a will to work hard during crunch times, your people skills are going to be as important as your technological knowledge, sometimes even more so. You're going to be meeting new people all the time, and you need to be able to work with them without much fuss. Even if you're the most gifted engineer involved in the project, if you cannot play at least somewhat nice with the other folks on the team, you will be utterly useless to your company. In short: You will have no merit. The project will be far better off (and more successful) by hiring somebody who's maybe not quite as brilliant on a purely technical level, but can actually work with other humans. A meritocratic system based purely on technical aspects (for example, code quality, whatever that may be) would fail to take that into account. Another big thing is politics. Somebody has to secure funding for our engineering projects, either from within our company, or from outside. They need to be both knowledgeable enough on a technical level to know what can and cannot be done, and maybe even a rough idea of how, and they need good salesmanship. Often, this is a senior engineer (at least in the teams I've seen so far). They don't need to be super talented on a technical level -- they might have written rather crappy code when they were younger. But they're absolutely crucial in the grand scheme of things. So for me, any CoC where they mention a meritocratic system is a pretty big red flag, to be honest.
  6. alpenwasser

    LTT Storage Rankings

    Just a general heads-up: I was abroad doing a crapload of overtime during spring, which meant basically no private live in the past few months. The good news is that I get to take lots of time off during the second half of the year, so I was planning to get around to this in the coming weeks. Nice to discover somebody has decided to help out! Many thanks to @timdine (and, uhm, apologies for the messy code ).
  7. alpenwasser

    Is Tesla the "Apple " of cars?

    Oh lala, people. We already have GPU and CPU wars, let's not add this one to the list, shall we? I've done some slight cleanup in the thread. Nothing personal, please keep it friendly. Thanks.
  8. That does sound pretty freaking funky. I must admit I can't really help much with Windows -- I'm a Linux guy, mostly. It does seem a bit weird that these issues started right when the electrician did his thing, but it could always be coincidence, of course (or just something really weird).
  9. alpenwasser

    AMD or Nvidia

    While we realize that this is to many people one of the great existential questions of life, given that we have had many, many, many flamewars due to it, I'm going to go ahead and lock the thread before things get out of hand (again). Thanks for understanding.
  10. If you're still having issues, I would indeed suspect that something is physically broken. Reflashing with the same version sounds like the right idea, though you could always try another version just to see what happens. I would try to make sure that it really is the card and not the drives or anything else in the PC before you go and buy a new one, but I realize that if you don't have another machine available to test things, this will be rather difficult.
  11. Power losses are tricky beasts. Usually nothing terrible happens, but it could be that some piece of hardware actually got damaged. Reflashing doesn't sound like the stupidest idea though; could be that something in the card's bios or firmware got corrupted but the hardware itself is fine. I'd also try to make sure that the HDDs themselves are still okay (I would presume that they are, but one never knows). And maybe try out some other storage device on the LSI card; something which hasn't had contact with it yet. If you do, make sure (if you can) that you've tested the HDDs first and that the LSI card doesn't kill stuff which is connected to it (unlikely, but I have actually had that happen with a SATA controller which was defective back in the day -- fried three HDDs of mine). Running the card in another computer could also be useful, if you can find one into which you can put it. A diagnostics utility could also be of potential use. See if lsiutil gives you anything useful. Zip archive download with versions for major operating systems here: https://www.broadcom.com/site-search?q=lsiutil It even has an efi version if you can't get into your OS; never tried that though, only the Linux version. Make sure to be careful; you can do pretty much anything with it, including breaking stuff. Here's some guidance on the tool: https://www.thomas-krenn.com/de/wikiDE/images/4/44/Lsi_userguide_2006_20130528.pdf
  12. alpenwasser

    LTT Storage Rankings

    Ah yes, that bitter cost of sweet, sweet redundancy. Been there, I empathize. I'll update the rankings as soon as I've worked out the kinks in the script.
  13. alpenwasser

    WD80EFAX-68LHPN0 - WD Red?

    Terribly sorry, but I've been a bit out of the loop and haven't had time to keep up with hardware, so any answer I'd give you here would be mostly guesswork.
  14. alpenwasser

    LTT Storage Rankings

    I see you've uploaded them to LTT now. I'd actually forgotten that you could do that (just goes to show how long I've been absent ). Yeah, Caselabs almost certainly have something which can accommodate 26 drives. Ain't going to be cheap though. You could also look for something with lots of 5.25" bays, and then fill those with hotswap cages. Given that some of your drives are 2.5", something like the Xigmatek Elysium could accomodate your HDD combo: - one 8 × 2.5" -> 2 × 5.25" bay adapter - one 4 × 2.5" -> 1 × 5.25" bay adapter - three 5 × 3.5" -> 3 × 5.25" bay adapters This would give you 12 2.5" slots and 15 3.5" slots, but it would be neither pretty, nor quiet, nor cheap. So I'd probably still recommend Caselabs. Or a custom mod, obviously, but that's going to be quite a bit of work if you want it done cleanly (it would fit into my server's case for example, and that's not even a huge tower). All working for me now.
×