Posts by ElReg!comments!Pierre 2711 publicly visible posts • joined 22 Jun 2009
> Pedant's corner, I know, but it's actually...
Nonono, it's the new spelling, all the rage on TwitBook this week, where have you been, like, in a cave on Mars or something? It's going to be in the dict* next month.
Also, in proper English "spell" is an irregular verb. Pedant right back at you!
*and this one the month after
I spot a textbook a posteriory selective factpicking. There are a few things to consider here:
Shakespeare had quite an aura, and quite a large audience, including most of the English-speaking "intellectual elite" of the time, he was not some 12yo posting to their friends.
Yet all these words did not make it to the dictionnaries until several years after he "invented" them (note that he actually did not invent most of them of course; he's just the first "tier-1" record of their use that we can find nowadays. They might have been common in the suburbs of London for several years before that, for all we know).
And finally if you pore over Shakespeare's writings you'll notice quite a few strange words that are understandable but did not make it to the dictionnaries of today (or only the most extensive ones, which mention such words _just_ because you may encounter them in a Shakespeare play).
Not every meme is worthy of dictionnary inclusion just because "Shakespeare Invented Words".
I'll stick with Baeckeoffe, thank you very much!
It looks like the Collins is trying to turn itself in the Urban Dictionnary. Are they also replacing all their examples with ones questioning the sexual orientation of some Oregon middle-schooler's classmates?
My sister is consistently refereing to blenders as "mix-your-soup" and it's catching up in her circle of friends. To be included in the Collins next year I believe, together with "twitbooking" (TBD; reposting of content cross-websites perhaps?), "connectidate" (TBD, probably meeting people online or sumfin), etc...
Language evolves, I agree, but including in dictionnaries what is nothing more than the "adorkable" portmanteau-of-the-month for a subpopulation of American teenagers is hardly evolution. I bet 2/3 of these words have only ever been used by a few thousands of people at a time and will stop being used at all before the print version of the dictionnary come out.
I'm sorry, none of this article makes real sense to me; I'm probably not familiar enough with Cisco's "web, email and content security management appliances" (and I thank Dog everyday for that fact, too. It's the little things in life, you know).
From the BSD bug report:
II. Problem Description
When an encryption key is supplied via the TELNET protocol, its length
is not validated before the key is copied into a fixed-size buffer.
III. Impact
An attacker who can connect to the telnetd daemon can execute arbitrary
code with the privileges of the daemon (which is usually the "root"
superuser).
Now, that I can understand. Pretty simple to patch really, unless I'm missing something. (also, ssh, d'uh)
Riddle me this -- why would software need any changes whatsoever for systemd, when systemd is just supposed to affect the bootup process?
Ha, but that's the point. systemd is not just an init system, it aims at becoming the entire system. It has it's own re-implementation of rather a lot of base GNU utils, for the sake of low PID count as I gather. Rather idiotic and dangerous for no good reason if you ask me, especially as the whole clusterfuck doesn't actually work very well -perhaps unsurprisingly.
But be aware, you will be alone in the future!
Yup, that's totally what happened when OpenOffice was forked to LibreOffice for example...
If there is a systemd-less fork, all the sysadmins will switch to it in a split second. We shall see then how desktop monkeys manage to maintain a huge distro on their own, and who will feel alone. I'm sure we'll miss you and your m4dz sk1775 immensely.
Hopefully this will go somewhere and kill the nonsense that is systemd before it ends linux.
Oh, systemd won't end Linux, it will always need the kernel. It may however end GNU (that's the long-term goal when you listen to Lennard P.)
Actually I'm half-surprised that Stallman hasn't said anything yet...
It works okay for desktop, but you'll see odd things on servers that have service quirks (the very thing systemd is supposed to make easy...)
I've seen intermittent (like, every other boot) problems with laptops too, mostly trackpad and/or WiFi chipset*. So much for the "faster boot" argument when you have to reboot immediatly after every other boot...
*problem gone after a revert to sysvinit, so definitely a systemd issue
Spot on Raumkraut.
There's also the little detail that the permission tree in systemd ("the user doesn't own the process, each process owns its children") is making a fragging permission mess very quickly which means that after a while you have to run a whole lot of shit as root, à la Windows pre-RT, or reboot to clean the mess up.
Systemd is a toy, not a proper tool. I say burn it. Burn it with fire.
Yup, same here. Go ahead and get rid of that systemd shit. I sure did go back to sysv on all my debian machines, systemd is just... not up to the job on a real machine. I can see how it can be useful on a "reboot-every minute" tablet, but on a proper computer it just doesn't work (on top of slapping all the UNIX principles in the face, repeteadly)
It's not like you'd need custom-build hardware for a TOR router. I reckon I could build one out of my Fonera in a couple hours, probably on top of a hardened realtime-patched minimal Linux kernel, a minimal set of system utils (busybox?) and a purpose-compiled TOR stack.
I would build it on top of Gnu HURD but it'd probably take a couple years instead of hours ;-)
So, unless they're selling it for its weight of SD cards filled with iTunes-bought songs* because they custom-built it, I don't see the problem. Oh wait.
*https://what-if.xkcd.com/108/
Actually a mini projector is not going to be much better than this. For anything good you have to cross the kilogram barrier. I should know, I tested everything I could get my mitts on. I finally bought a LED thing, on the light-and-small side of the large projector range. Pretty quiet too, and not overly expensive.
At first I was more interested in a mini projector because, well, neato! But I found even the rather overexpensive ones struggled at more than ~1.5 m, and that's before we even start talking contrast. This embedded one is probably not extra-good but certainly worth having for a quick improvised meeting on the go where the mini projector is going to be equally shitty with the added burden of needing an extra couple cables*, an extra device, and an extra power outlet.
*and hdmi cables are not the sturdiest things around, either. Being carried around will mess them up pretty fast.
> but it's surely a fine principle?
It certainly is, but that's where the trouble kick in regarding gender equality: for the end of the evening shift, say, starting at half past pissed, you may want to have more bartenders of the burly, hairy-chested type as opposed to the squishy curvy type, for obvious reasons (including the squishy type not wanting to actually get squished)... and you're not paying the same wages... see the problem here?
Of course there's a very valid reason to begin with, but from outside it may look like you're willingly paying women less than men. That's where you have to put in all these compensatory-this and equivalent-that which makes the calculations that much more complicated, as I was saying (although I did not have this particular problem in mind at the time, I was thinking more along the line of out-of-hours work and such).
> 2 people, doing the same job, will be on the same pay-scale, at points determined by their respective experience/performance.
Yes, they are on the same payscale. It would be illegal to have separate ones. The point being argued here is, is the "fluffy factor" (experience/performance/etc recognition) evaluated and taken into account the same way. And that's where the icon is NOT sarcastic.
> "I believe men and women should get equal pay for equal work. "
> I believe that it's insane to suggest anything else.
Sure, but in the real world how do you do that? First there's a problem with the metrics. Do you consider yearly pay, per-hour, bi-monthly, on the results (per-project)? With each method comes caveats, as not everyone works the same number of hours, results can depend on your work and abilities but also on a good serving of luck, etc. Then there's the fact that in a lot of places pay calculation depends on your overall experience and on your seniority in the place, meaning that things get hairy pretty quick if you're trying to compare pay between 2 people of different ages and/or who weren't hired at the same time. So the only way really is to make it retroactive, i.e. in the end of the, say, year, you crunch the numbers, with a lot of equivalent-this and compensatory-that (to account for holidays, number of worked hours, position in the local food chain, performance reviews etc) and you can tell if you've been paying fat black redhead women less than fit blond white men (or any other 2 categories that you may want to compare).
Then you can retroactively make adjustment "attention all blue-eyed staff, please head to your local HR office immediately for a 2% pay cut. Thank you for your cooperation" or sumfin' (OK, not really, what you would do is tweak the raises for the upcoming year I guess, and hope that things will equilibrate).
Not as easy-peasy as it may seem at first.
You're being a bit unfair there. He's just complaining that Intel oversold the precision. He did not seem so upset about the precision itself, but about the fact that if you rely on the doc you'll assume that your result is more accurate than it actually is, and thus rely on it instead of doing things another way.
I bet he's perfectly happy with Intel's response of fixing the doc to fit the code.
Personally I don't understand the total dislike for it that some people have but then I am quite happy to see people build distros that use SysVinit if that's what they want. There is room for both camps.
You'd think that, but actually there isn't room for both camps when one camp is trying to push kernel commits that would break the other camps' solution.
What I dislike in systemd is not primarily what it is* but more what it wants to become: OS in place of the OS (that's the self-admitted goal of Poettering, for whom systemd is poised to become "a group of blocks from which to build an OS" -or something to that effect). It means that devs are faced with a choice: either you code against systemd, and then you let it do everything it does because redundancy is not good (and in some cases even impossible); that means removing key parts of other userland tools, and even modifying the kernel in some cases. That also means that other, simpler init methods won't work anymore. Or you let systemd alone entirely.
*although I do think that journald's binary logs and the all-encompassing, everything-but-the-kitche-sink approach to software design are imbecilic, to name only 2, but that's only my opinion.
On the last part of my post I have an example: I usually try to minimize unnecessary resource hogs so I use minimal windows managers, like twm, quite a lot. twm windows don't have a "close" button: it's a widows manager, it gives the application a window to work in, that's it that's all. BUT most windows managers do have a "close" button... and although every sane developper includes one, sometimes several, in-app ways to close said app, one particular project visibly thought "windows manager have a close button anyway so why bother" and did not provide any way to cleanly close the app from within.
That -otherwise excellent- application is Rawtherapee (give it a try, it's not quite as complete as lightroom but it's stable, has everything *I* need and then some, and is open source). Much to my dismay, to close it nicely in twm you have to send it a kill -15 signal from a virtual console. It's not lazyness, as the application itself is extremely well finished, has a lot of well thought-out functionalities, and very complex ones to code, too. The app's worth it but it's annoying, and a sign of what happens when you start relying on "gadget" niceties in parts of the system that don't require them.
> not to be picky, but could you be more specific about pulseaudio?
I can answer only for myself, and perhaps it's fixed now, but until recently it kept crashing on me, repeatedly, often bringing the calling app down with it. Be it for games, online video, local video, anything a bit demanding that had sound was likely to crash within a few tens of seconds (with an error message about pulseaudio on stdout). That, or grind to a halt with distant "tt tt tt" as the sole audio output (and a whole bunch of pulseaudio error messages to stdout). That on an otherwise remarkably stable system. Pulseaudio had become synonymous of crash: whenever anything crashed or stalled, you could be sure there was a pulseaudio error message waiting for you on stdout. Switched back to a more "basic" sound system and everything magically works.
It may be because I'm always using bare-minimum hardware configurations, perhaps a few more GB of RAM would have solved the problem, but if so it's still bad code IMHO.
> One of the MASSIVE benefits of PA is that EVERY process can be given its own volume control.
You see it as a good thing, I see it as a telltale sign of crud. Typically that's not the job of the sound daemon, that's the app's job. Same as how systemd is trying to do everything at once. Not only is it bad design because it leads to unmaintainably huge codebase and some features will inevitably fall behind in terms of development, it also means to encorage app developpers to not include volume control even though their app would need it -because it's done by the sound system anyway so why bother, right?
Why a reinstall? I just switched back to sysvinit, no problem at all (I did have to remove a few things, mostly meta-packages with one "gadget" dependency coded against systemd, but no core thing).
I too was disturbed by Debian's switch to systemd. That's only for the i86 version but i reckon it's the most used one. Dangerous to say the least.
> Pulseaudio and avahi are garbage as well.
Tru dat. Perhaps I failed to throw enough resources at them, but I never could make them work reliably.
> He singlehandedly is the reason I won't have a Linux Desktop any more.
He singlehandedly is the reason why I have Linux desktops that don't have either Avahi or PulseAudio. Or systemd for that matter (tried it, reversed to sysV after a couple magistral borks on update)
> that is no different on windows forums, though.
It is a bit different on Windows forums for 2 reasons that Windows transfuges often don't get at first when posting on *NIX forums:
-on widows forums you usually talk to fellow users; it is customary to begin with "This app is crap and the doc is useless, please help a bro out". On Linux forums you're usually talking to the people who coded the damn thing and / or wrote the doc. You can't be a total dickhead to their face and demand that they fix your problems politely.
-in the Windows ecosystem when you talk to the official support and/or the devs they already got your money. You're a customer and thus benefit from a higher tolerance if you're being a bit of a dick. On linux forums (excluding the paid-for support like RHEL) you usually did not pay the devs you're insulting, so they don't have to be nice if you're being a jackass.
Now write me a detailed tutorial on how to do $STUFF because I can't be arsed to read the man page or search the forum/mailing list history.
Oh, and did I mention that your stuff is not quite as good as anything else on the market and the only reason I try it is because I can have it for free?
I can sort of believe it...
> That fanboi was almost certainly correct. Strictly speaking, no virus has ever been found running on a Mac.
Since you're going for the pedant angle I thought I'd mention that there has indeed been viruses developped for the Mac; including the first virus ever to be described as such, actually. So, strictly speaking as you would say, viruses were born on a Mac.
Now please by all means do carry on flaming each other over how such or such OS is totally foolproof, I don't really care about OS wars.
> Yes it is. Mostly because it's the same chip with the same interface.
Neither the same chip nor the same interface actually. And that's part of the reason why it's not (even remotely) as fast. But again, it doesn't really matter for the expected uses of a smartphone, except perhaps way-too-high-pixel-count video shooting.
"Is an SD card as quick as built-in memory? I assume it is, if you get a decent card?"
Not even remotely close, but a decent speed class card is usually fast enough that you won't notice the difference on a phone, unless perhaps if shooting insanely high-square-pixel video directly to card.
AFAIK as a rule of thumb the speed for such devices is internal flash>CF card~USB drive*>>SD card
The internal memory of the iPhone is compact flash, and with a high-speed interface I would expect, so it's expected to be massively faster than a SD card. By a factor 10 at the very least I would guess. Again, for most uses it probably won't be noticeable very often especially on a phone. For example my Pi can stream 720p video seamlessly from a relatively low-speed (class4) SD card.
*both USB devices and CF cards speeds vary enormously depending on the interface, the storage hardware and the iteration, so they leapfrog each other on a regular basis; they are not usually direct competitors though, so the comparison is not really relevant)
Not taking any side here, but the strong selling point of the iPhone is its luxury feel, coming from the use of materials that feel sturdy. That is exemplified time and time again when fanbois deride the competition: the first argument they use is usually "cheap plastic feel".
Thing is, when you drop a "cheap plastic" shell, its deformation will absorb and spread a lot of the shock and keep the rigid screen relatively protected. The case might also get a dent in the process, but usually not even. The (more rigid) metal back may pop off if present, but pick the thing from the ground, reassemble as needed and you're good to go.
Now when you drop a sturdy-feely rigid case it won't deform (it may dent but that doesn't dissipate nearly as much energy). So the shock is transmitted in full to the gorilla glass, which is many things but not flexible. >SCHATTARZZZ<
If Apple's PR dept blew a fuse over "snow lepperd", I don't think "fruity führer" will do their arteries any good.
But El Reg now has considerably more clout than back then; it's recognised as one of the leading tech publications in the English*-speaking world. So perhaps Apple will be willing to let a few disparaging comments slip through...
*English and related languages such as Strine or Murkin.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
