Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!


Forgot your password?

Slashdot videos: Now with more Slashdot!

  • View

  • Discuss

  • Share

We've improved Slashdot's video section; now you can view our video interviews, product close-ups and site visits with all the usual Slashdot options to comment, share, etc. No more walled garden! It's a work in progress -- we hope you'll check it out (Learn more about the recent updates).


Comment: Re:IBM should put SCO out of misery (Score 5, Informative) 170

by Damouze (#49283287) Attached to: Not Quite Dead: SCO Linux Suit Against IBM Stirs In Utah

The SCO that is currently trolling IBM is not the SCO that you remember as "such a good company". There are two SCO's:

* The Santa Cruz Operation (1979-2001). This is the SCO that you remember. They brought us Xenix (bought from M$), SCO Unix and Unixware. This SCO sold their rights to UNIX to Caldera Systems (then primarily known for Caldera/Open Linux and OpenDOS (bought from Novell, which had in turn bought it from Digital Research earlier). In those years they were mostly famous for filing an antitrust campaign against Microsoft). After selling their UNIX servers and services division to Caldera they renamed themselves as Tarantella Inc., after the product line they retained. Tarantella was subsequently bought by Sun Microsystems in 2005, which in turn was bought by Oracle in 2010.
* The SCO Group (2005-), formerly known as Caldera Systems / Caldera International. As Caldera they bought above SCO's UNIX servers and services division and subsequently renamed themselves to "The SCO Group". Like an evil David they tried to topple Goliath IBM by (falsely) claiming in court that programmers from IBM illegally copied code from SCO's OpenServer sources (supposedly their intellectual property was so secret that their allegations of verbatim copying code by IBM was "proven" by presentational slides which had the SCO code shown in Greek alphabet). Around the same time they started selling subscription based Linux licenses to large IT companies (which were led to believe that The SCO Group owned the rights to Linux). This ridiculous scheme went on for several years, until a judge decided, once and for all, that enough was enough and told them to bugger off, as in the meantime, it had become clear in a separate lawsuit that Novell was in fact the owner of the UNIX copyrights, not the SCO Group.

Comment: Capital punishment is NOT humane! (Score 0) 1081

by Damouze (#49259909) Attached to: How To Execute People In the 21st Century

The driving force behind the creation and abandonment of execution methods is the constant search for a humane means of taking a human life.

Capital punishment is not humane. There is no humane way to take a human life. That is a contradictio in terminis.

The death penalty should be abolished worldwide as soon as possible. Aside from the fact that it is barbaric and has no place in a truly civilized society (to the extent that no nation or society can really call itself civilized, unless it has abolished the death penalty), there is always a chance, however slim, that the person being executed is innocent of the crime he was convicted for. Someone who has been put to death cannot be brought back to life if he turns out to be innocent. A person who has been sentenced to life in prison, on the other hand, can be released and rehabilitated.

Comment: Re:So roll your own. (Score 1, Insightful) 716

by Damouze (#49045715) Attached to: Is Modern Linux Becoming Too Complex?

Systemd is a monstrosity that should never have seen the light of day. There is no excuse for closed-mindedness, arrogance. They lead to bad design. However brilliant its creator might be, systemd is flawed by design.

Some highlights (or maybe 'lowlights'?):

* System logs are primarily kept in a binary format, which is a very, -very- bad idea. You wish to store them in a database (which is basically what happens)? Fine, but only as a secondary way to store them, not the other way around!
* Init was built to do one thing and it was built to do that thing very well: run some variant of a shell script called rc. Everything else is superfluous. It is the cardinal rule, the core philosophy UNIX (and Linux nowadays) was created upon.
* Systemd is not very transparent. That in and of itself makes it a security concern to say the least.
* Nearly everything with systemd is done through (binary) modules with XML configuration files. A bad idea. A very, -very- bad idea. The pros of using shell scripts, and maybe, maybe use some (human-readabe) external configuration scripts are many, not the least of which is the fact that they are by far, much more customizable. With a classic init setup you not only control the setup and configuration of your system, you actually control its behaviour at boot time as well. If there is something you don't like about your system's behaviour, all you have to do is modify one or more of those shell scripts.

Your notion that systemd finally puts the dynamic parts in user space does not track at all. The dynamic parts have always (also) been in user space. That is why tools such as ifconfig, route (combined in the iptools ip* commands nowadays), etc. are user space / user land tools; they do not run in kernel space. The kernel provides an API for user space tools to work with. It has done that since the very early Linux days. The rc scripts and the scripts in init.d (in a SYSV like Linux setup) or in rc.d (in a BSD-like Linux setup, not many of those around anymore) in turn then use these userland tools to setup and configure your system.

If your network interface comes up too late during boot, before something that depends on it comes up, the problem is with the boot sequence, not with init itself. It's something you can fix yourself. You don't need some black box to do that for you. Moreover, if it succeeds at all at doing it, it will do it very badly. There is no excuse for a bad system setup. That goes both for the maintainers of a software distribution (in this case Linux) and you yourself as the individual who operates that software.

One thing I do agree with you about is the serious case of bad documentation. Moroever, I think that bad, incomplete or even missing documentation is the source of many a Linux user or admin's woes. But it is something that everyone can contribute to. Linux is after all, Open Source.

Comment: Re:Quebec Language Police (Score 1) 578

by Damouze (#48725843) Attached to: What Language Will the World Speak In 2115?

Your argument is flawed.

Technical and scientific jargon is usually derived from Latin, Greek or both. That is why the word electricity is so familiar to French, German and Dutch native speakers. The word was derived into each language from Latin.

If I put the word through a mangler and try to write it as the Romans would have, we would end up with electricitas, with an oblique stem electricitat-. Since a lot of words originally from Latin (and I am not arguing that word electricity is actually one of them, just that it was derived from a Latin root in an analogical fashion), find their reflections in later languages in their oblique form (we see this in the plural forms of western romance language nouns).

A system of derivation - either natural or artificial - tends to follow a certain paradigm and the odd ones out tend to be sucked into the paradigm as well (Icelandic has a few nice examples of this as well if I'm not mistaken) if their base forms resemble it enough. English has words like these: fish (singular), fish (plural), but also fishes (plural) in certain dialects. Some English verbs that used to show vestiges of the old Germanic strong verb classes have transformed into regular ones: as well, as did some verbs that fall into a category called preterite-present verb (which is a verb that has a present tense meaning, but a conjugation in a preterite (simple past) form. Their past tense was originally formed regularly according to the Germanic weak verb system. In English they are more commonly known as modal verbs.

The verb "to owe" for example is originally a preterite-present verb. The only remannt of that preterite-present past is well, its past tense form "ought". While its original meaning was to indicate one's posession of an object, along with the transformation from preterite-present verb to regular Germanic weak verb, its meaning shifted as well, to "to be in debt", with the regular verb "to own" taking over its original meaning.

Okay, I got a little carried away... Back to the topic at hand.

Like I said, there is a large chance that a word from scientific jargon was originally derived from a Latin or Greek word, or a compound of both. The chance that their reflections in other languages will be derived from that same base word is large as well.

Comment: Re:Call me racist and evil and bigoted and everyth (Score 1) 158

by Damouze (#48615399) Attached to: 9th Circuit Will Revisit "Innocence of Muslims" Takedown Order

Thought and action are to different things. Speaking one's mind is acting on one's thoughts, just as much as otherwise acting out on one's thoughts and urges would be. From there on one enters the big grey area where a legislator could potentially infringe upon one's basic freedoms. Potentially being the operative word here. But the opposite is true as well - by speaking one's mind one could potentially infringe upon the basic freedoms of others as well. One's own freedom ends where those of others begin and vice versa.

Comment: Call me racist and evil and bigoted and everything (Score 5, Insightful) 158

by Damouze (#48599423) Attached to: 9th Circuit Will Revisit "Innocence of Muslims" Takedown Order

A person who has not commited a crime is innocent.
A person who is accused of committing a crime is innocent until proven otherwise

That person's beliefs, religious or otherwise are in and of itself not criminal. It is what that person does with those beliefs that makes all the difference.

Comment: Re:Trying to wrap my head around this (Score 1) 389

If you are against copyright, you want the terrorists to win. Please don't try wrapping your head around this yourself: highly skilled congress members are already wrapping their hands around this for you.

This posting was paid for by the RIAA and MPAA.

I assume you're joking here, because that is literally the dumbest remark about copyright and terrorism that I've ever heard.

Comment: Re: are the debian support forums down? (Score 1, Informative) 286

by Damouze (#48179339) Attached to: Ask Slashdot: Stop PulseAudio From Changing Sound Settings?

Skype on Android has this nasty "bug" (although people might be tempted to call it a feature): you can't really close it. Believe me, I have tried. I have all but given up on it. Even if you have all the notifications for it disabled and forcibly stop it, it will respawn automatically within a show period of time. Were it not for the fact that I need it for work, I would not allow an app like that to even be installed on my cellphone or my tablet.

With regard to systemd: if the only choices you have are broken ones, you don't really have a choice now, do you? On the one hand you have systemd, which is bloated beyond repair. On the other hand you have all those software packages that won't run otherwise because they depend on it.


Comment: Re:Boycotting RHEL7's uselessd (Score 1) 469

by Damouze (#47960763) Attached to: Fork of Systemd Leads To Lightweight Uselessd

I have no idea why Redhat made so many changes in their most recent release, but it is so vast that it may as well be a completely new distro. To name a FEW:

  Anaconda RHEL installer completely redesigned

About time.

Legacy GRUB boot loader replaced by GRUB2

Adds a bit of complexity to it, but GRUB2 is much more versatile than old-fashioned GRUB. Besides, it's also much more mature now.

Procedure for bypassing root password prompt at boot completely different

As long as [sudo] su - still works (with any kind of password), I'm happy. It's root. You're not supposed to bypass the password prompt! But, if you really, really, want to, you could always issue init=/bin/bash at the kernel command line in grub[2]. Used to work with lilo, still works with grub[2] as well.

SysV init system and all related tools replaced by systemd

The sheer horror. Seriously, another reason NOT to use Red Shit.

ext4 replaced by xfs as default filesystem type

That, at least, is an improvement. Both ext3 and ext4 have fundamental design flaws (like kjournald (and alike) that pops up every five seconds and slows down your system to a grinding halt if you're especially unlucky because it fails to check if it's already running). XFS is a much more robust design in any case, and way, way faster to boot!

Directories /bin, /sbin, /liband /lib64are now all under the /usrdirectory

Not too sure what to think of this. On the other hand, Solaris does basically the same and has been doing that for a quite while.

Network interfaces have a new naming scheme based on physical device location (e.g., eth0might become enp0s3)

Not sure what to think of this either. On the other hand, various UNIX variants do more or less the same.

ntpdreplaced by chronydas the default network time protocol daemon

Hmm. Not familiar with that one. Is that the one that will absolutely refuse to update your time and date completely?

GNOME2 replaced by GNOME3 as default desktop environment

Arf. Gnome. Nuff said.

System registration and subscription now handled exclusively with Red Hat Subscription Management (RHSM)

I blame Oracle for that.

MySQL replaced by Mariadb

I blame Oracle for that too.

tgtdreplaced by targetcli

What a shame. Last I checked, tgtd was just about the only ISCSI target daemon that made any kind of sense. But I admit, it's been a while.

High Availability Add-On: RGManager removed as resource-management option (in favor of Pacemaker)

Just a symptom. WIth systemd as the beating heart of your system, you'll need that Pacemaker. Especially if it's in the death throes of the log corruption you are bound to get.

ifconfigand routecommands are further deprecated in favor of ip

I love the 'ip' command. It's powerful. Still, for heaven's sake, let's keep the old commands?

netstatfurther deprecated in favor of ss

Stupid decision. Netstat is at the core of many a UNIX admin's skills.

System user UID range extended from 0-499 to 0-999

Couldn't care less. As long as I can reserve 1000 for myself (pun intended).

locateno longer available by default; (available as mlocatepackage)

Hmm. Why on earth would one want to do that?

nc(netcat) replaced by nmap-ncat

Well, nmap is a powerful tool. This, for once, makes sense to me.

Systemd is pain to use for me and feels backwards... I find troubleshooting processes with it to be more frustrating than anything else Redhat has done in the past 20 years... Well, almost.

More frustrating than using what's basically a GUI-only installer on a system that really does not need a GUI in the first place? Sure, you can use the anaconda text mode installer, but it does not support LVM and if there's one thing you really like on any Linux system, it's LVM. It's clean, it's mean and it's easy to setup and use. Actually it's easier to setup outside any installer, but anaconda (at least in text mode) refuses to pick it up and will happily propose (and if you let it, setup) a non-LVm installation for you anyway.

Comment: Re:Funny inability to see alternatives (Score 1) 469

by Damouze (#47960527) Attached to: Fork of Systemd Leads To Lightweight Uselessd

You know. I could not care less about systemd, journald or wayland.

Wayland is just a piece of crap that supposedly should bring the desktop to Linux. Well, here's a reminder: the 'desktop' has been available for Unix like operating systems for the past 30 years or so. It's called X. But the developers of Wayland (or Mir) don't like X. The fact that it is a f****ng UNIX standard does not even come to mind! Instead, they decide to reinvent the wheel. Twice! But... Remote X sessions? No way José!

I have played around with systemd and journald. It's sort of fun. Until you realize it breaks the very thing that it is supposed to provide: a standardized way of booting up your system. Again, someone tried to reinvent the wheel. And now twice as well!

And why the hell do I need dbus? Come on, can't people invent an IPC mechanism that is even marginally more useful than that and at least more well-programmed and well-behaved? What do I need dbus for if all I am doing with my system is say, running sendmail?

Anyway, enough with the ranting. Uselessd is a fitting name. Even so, adopting it (or systemd) requires a change of philosophy, one that I am not willing to make. Linux (and UNIX in general) is supposed to be an open system with a intelligible interface. Hell, all init is supposed to do is run a shell script! It is not supposed to be this big binary blob that only takes up memory. Memory that I could be using for other things, like say, run sendmail...

You will have a head crash on your private pack.