Geoff Huston wrote an article about comparing 6to4 to IPv4 in terms of failure rate - and found out that about 0.2% of IPv4 connections to his web server were also broken. Geoff's article also provides insight why exactly a large percentage of 6to4 connections to his web server failed: routing packets around the planet just because a lack of 6to4 gateways and in three out of four issues, some broken firewall dropping 6to4 packets.
Issues A and B also doesn't necessarily mean that IPv4 is permanently broken but "occasionally". For example, every mobile carrier in my country deploys Large Scale NAT/Carrier Grade NAT for IPv4, but in order to max out those boxes, they're running with very low session timeout settings. Their NAT routers silently drop my session when a tcp connection is idle for longer than a few seconds. While web browsing "usually" works, things like IMAP sessions very often do break and reconnect, for interactive use I'm forced to run ssh-sessions with "ServerAliveInterval 7". One of those carriers even temporarily blocks access to Apple's iTunes store - maybe because the iTunes store is known to eat up to 300 parallel NAT sessions for a single user (compared to roughly 20-30 for "usual" web surfing). When accessing some very slow web server, the NAT session timeout also kicks in, resulting in my browser "endlessly" loading the same page.
Right now, the same carriers don't yet offer IPv6, so technically, they're forcing me to issue "A". Once they do offer IPv6, my mobile internet access is likely to be issue "B".
The same experiment can actually work out very differently. At Google's IPv6 implementors conference in summer 2010, a japanese ISP reflected about the very same experiment like heise.de or the World IPv6 day do of adding AAAA-records for a day. They've been doing IPv6 for years now, including hosting via IPv6. When they added AAAA-records for their very large japanese portal site biglobe.ne.jp, they lost about 5% of page views immediately and 5 minutes later, their phone started ringing endlessly. A few hours later, they've chosen to cancel the experiment by removing AAAA-records from their DNS. In my mind, many japanese ISPs have been using and offering IPv6 access for years now, but there haven't been any major services available via IPv6 in Japan, so the actual IPv6 traffic has been very low and most people weren't aware that their IPv6 setup is simply broken. Maybe even Yahoo's and Google's often-quoted "0.025% of users do have IPv6 issues" bases on Japan being largely broken in terms of IPv6 service while the rest of the world may run IPv6 without any issues:-)
Well, Germany is quite a very different issue. Most large german access (DSL/broadband/dialup) ISPs don't yet support IPv6 and the de-facto standard-dsl-router range of most ISPs (AVM's Fr!tz-box) didn't support any kind of IPv6 at all until quite recently. Even now, IPv6 is something hidden deep in their menues and actually needs to be manually turned on. German web hosting consists of a few large companies, where support of IPv6 is currently left as a DIY-option for dedicated servers and not for any shared hosting plans. On the other hand, close to every ISP peers via IPv6, is running 6to4 gateways and happily runs IPv6 on their own networks, but IPv6 isn't yet used for any actual major public service, so in theory, IPv6 shouldn't be that hard to get working in Germany today... but for today, IPv6 in Germany is actually VERY poor.
To illustrate how worse IPv6 in germany is, check the TLD stats at Hurricane Electric, compare the amount of AAAA-records vs. the amount of A-records. For about every TLD (.com,.net,.org,...), there's one AAAA-record for roughly about every 90 A-Records. For.de, only about one out of thousand A-Records do have an AAAA-record. That's a ten-fold in being worse!
So heise.de didn't really venture a lot when they turned on IPv6, as even far less users in Germany actually do use IPv6 than in any other country. However, they've still done something very intelligent: once German Internet access ISPs do turn on IPv6 connectivity for their customers and customers notice about heise.de being unreachable, heise.de users are already aware that heise.de has been served via IPv6 for months without any problems, so any brokenness must be related to their own ISP (or their personal setup). They'll directly complain to their ISP and won't blame heise.de.
In fact, there are quite a few people out there using Anycast for TCP-sessions. It's really a matter on what timescale you're looking at. The networking guys see TCP as something to use for long-living connections - e.g. a BGP session running for days, weeks or even months. A flapping route in this setup will result in a broken session. But: what does this really mean to you? If your CDN distributes downloads which are "done" within a few minutes, such a rarely flapping route will result in a few broken sessions once a day out of millions of downloads successfully served. Compared to issues like non-working DNS, overloaded servers and filled lines, that's nothing and can actually enhance the overall CDN service.
A nice paper to read is this one from Matt Levine. He's working for a CDN provider using TCP-Anycast for years now and sums up the most important issues on TCP-Anycast.
Basically the most important one is that your anycasted servers really have to be spread far enough so that flapping routes at some peering point won't matter. As a rule of thumb, put one CDN loadbalancer on the US east coast, one to the US west coast, another one to western europe, one to Australia and one in Hong Kong. If you'd like to put multiple CDN loadbalancers to one continent, leave space between them, e.g. one box for each country/state.
Until recently, Kindergartens weren't permitted at all to copy single song sheets, their only option were to buy the books of those song sheets, but those books again didn't permit copies at all. However, Copyright expires 75 years after death of the writer; and those songs may be freely copied.
Now, there's also the option for Kindergartens to pay a fee of 56 Euros (plus 7% VAT) per year (€44,80 for Kindergartens operated by churches or cities) which permits up to 500 copies of song sheets. For example, for 112 Euros (plus 7% VAT), up to 1000 copies are permitted.
The odd thing with this option is, that kindergarten teachers then do have to keep an account on the amount of copies being taken for songs whose writer is dead for at least 75 years: bureaucracy at its best.
Back in the "good ole days of the internet", IP-addresses were given out as permanent property and there's about no legal way for IANA or the current RIRs to recall those IP-adresses.
Nowadays, IP-addresses are given out as some kind of semi-permanent lease. For example, your RIR may offer you a larger allocation than a requested one, but they also may require you to hand back your old allocation after a few months to allow renumbering your old IP space.
Even forcing those organizations to hand out their/8 does give us about two more years, that's simply ridiculous. IPv6 has been in work since last century's nineties, experimental networks like 6bone were closed in 2006 after IPv6 has been declared as being stable for "production use".
Even older operating systems like Linux 2.4 and Windows 2003 have IPv6-stacks officially being seen as "stable for production use" (Windows 2003 is lacking IKE for IPsec, but that's all about it). Whoever isn't capable of deploying IPv6 within the next two years deserves being doomed.
IANA reserved 224.0.0.0/4 for "multicast"-usage, that's the equivalent of 16/8-sized networks. Renumber multicast!
We also need to replace RFC 1918's wasteful use of 10.0.0.0/8. No organization ever needs 16 Million IP addresses, even Google has a fraction of physical servers than this.
While I'm thinking about it: RFC 3330 spends more than 16 Million IP addresses for a single box. Whoever is using this 127.0.0.0/8: please do renumber to e.g.::1/128 and return 127.0.0.0/8!
I want to see what I'm working on and not have to deal with... my hand and wrist covering up my work.
A problem that utterly destroyed the work of amateurs like DaVinci, Michaelangelo, and Raphael, right?
During his last two decades, Ludwig van Beethoven lost his hearing. He was completely deaf when he composed his ninth symphony (famous for e.g. "Ode to Joy"). That doesn't mean that a hearing impairment enhances songwriter skills.
Imagine the works of those artists if they weren't bound to cover the area they're working on. They might've raised the bar of perceived perfection to even higher levels.
You said you've been using your 1&1 webspace for hosting a counterstrike mirror back in Germany during CeBIT 2000.
CeBIT 2000 took place in the end of february, so lets think about spring of 2000 back in Germany.
Back than, it has been quite hard to get anyone with decent admin knowledge on the job market, as the internet hype has just been in its hot time. So I'd expect any company to let their techs do some real work instead of hanging around at some trade fair booth, sipping coffee and chatting with some people passing by. I guess you've been talking either to sales or someone from the user helpdesk.
Back than, broadband was not really wide-spread in Germany - the very first lines DSL-were installed in the summer of 1999 in very few selected cities. Cable was not available at all in Germany and about everyone who hasn't been working at a university and didn't have a 2 mbit leased line in their company was connecting to the net via ISDN or modem. According to Wikipedia, only 2900 DSL lines were installed in Germany in back 1999.
ISDN was somehow spread, but access was billed per b-channel, so most users would only connect to their ISP at 64 kbit/s. Most users were using simple modems, connecting to their ISP at something around 40-52 kbit downstream (depending on line quality). So for most users, it took around 2-3 minutes per MB of download.
HTTP 1.1 (which enables pipelining and partitial downloads) was published in June of 1999. For "download sites" and "mirrors", you were expected to offer (anonymous) FTP, as http back than has been too unreliable and when your download aborted, you had to re-download the complete file.
Average workstations were running at 128-256 MB of RAM, the average server back than had something between 256 and 512 MB of RAM. Larger boxes were lucky of running at 768-1024 MB of RAM.
Back than, you could easily DoS most web servers just by opening a few hundred idle connections. Software back then wasn't really built to withstand any higher usage scenarios or serious DoS attacks.
Back then, people optimized their images by hand, cutting them into the magic 216-colour-"netscape"-palette for GIFs and made images small enough so that the web site would load within a few seconds.
1&1 is known for hosting sites on non-clustered, "smaller" servers. While upgrading their UPS in 2001, they experienced a major power outage, but were able to get back online within a few hours, as those hundreds of boxes did run their fsck in parallel. To compare: a few weeks later in 2001, their competitor Strato's highly-available storage clusters went offline for about a week and had to run a full fsck.
"1&1 Puretec" back than has been offering shared web hosting plans for hosting personal and small office websites, which translates to "small files, short-living http-requests". Like today, 1&1 does rely on the Apache web server. You had ftp-level access to your website, but your hosting IP address has been shared among dozens of other customers, so you didn't have some way to run anonymous ftp on well-known ports.
Apache back than also had the quite negative behaviour being served to a client to load about the whole file into RAM.
Counterstrike has always been one large file of around 70-120 MB in size.
Now imagine what might've happened. Got it? No?
You've been hosting some file in a country where the average user would need a few hours to retrieve that file from some service which has been built for about the exact opposite of what you're doing. Once just a few people try to get your "mirror" file, they'll bring the web server down, swapping and crawling on its knees.
Your website is hosted on the same ip address than hundreds of other customers, so your website can't be moved to a different server. About the only thing the admin can do in that situation is to get those hundreds of users other than you back online is either to swap the core component of web hosting software (Apache) by "something yet to be invented", change from a non-clustered hosting service to a large-scaled cluster or simply prevent users from downloading your file.
How much crime does a better passport stop, anyway?
It doesn't prevent any crimes and was never made to do so. A passport is just a commonly accepted item for identification.
In fact, many countries don't issue a passport unless you're applying for one, and the main reason for having a passport is the point that some country other than your own would like to verify wether the one who wants to pass their borders is the one claiming to do so while making sure that a blacklist of "known bad people" won't enter their country.
If you're reading the technical specs on the passport documents, you'll note that the biometric information (JPEG of photo, hashes of fingerprint marks) is stored into an RFID chip on the actual passport, and not somewhere else.
Even just the thought of inter-connecting the millions of passport-checking locations and granting those passport-checking devices (who are under control of hundreds of different countries with tens of thousands of IT operation teams) international access to distributed giant databases is ridiculous in the eyes of anyone who ever tried to set up a mashed VPN network.
In clear: there is no online comparison or verification of those biometric data against the data provided by the passport-issueing country. What actually may be verified is that a scanned image of the real photo matches the JPEG stored on the RFID chip and that the JPEG is cryptographically signed with a known-good signature.
And that's the point: the biometric data in an RFID chip are being used to aid against illegal passport duplication.
So in essence, the guys make copying a passport much harder. It does no longer only take some better colour printer, sealing those printouts in plastic and wrapping it between cardboard sheets, now it also needs someone who can break public key cryptography. That's all.
The 64-bit NTP timestamp spans 136 years with a resolution of 232 picoseconds, the 128-bit NTP timestamp spans 584 billion years with a resolution of.05 attoseconds - so right from those points, NTP is good enough for your applications.
What's still problematic is a problem that NTP also tries to compensate: the network latency. When you're receiving just two packets with exactly the same latency, you can't be sure that the third packet will be there with the same latency, so you're having an possible error rate of 33%. However, if you've seen a million packets with the same latency, your possible error rate is very close to zero, and that's why NTP can sort out the network latency problem only over time.
As forcing legalese through babelfish usually gives quite "strange" results, here's my abstract in short and layman's terms:
After 6 months of being employed in a company, you may ask your company 3 months in advance to a date set by your own that you'd like to cut down your worktime to some fixed amount (less than current). If company can't show any clear, legitimate reasons (e.g. extra-ordinary costs or security) why your job can't be done part-time one months in advance to your set date, you'll be working part-time.
An extra-bonus point applies for the employee, if the company has more than 15 "regular" workers: the act expects that such "large" companies are able to compensate part-time work.
Various non-discrimination rules do apply: your net income for e.g. a 20h/week job may not be less than 50% of the net income for the same week running at 40h/week. And once there are new fulltime job offers, those offers do have to be made available to current employes, where giving those jobs to part-time employees has to be preferred. Once you're working part-time, you may also apply for more work hours.
Well, the reasoning is simply that the update swapped some shared libraries, which are in use by more than a few known applications. Of course, Apple might say "Please restart at least your browser after updating and any other open application which somehow misbehaves", but many people do fail to do so ("well, I'll do it later" - and suddenly remember that thought once their browser plugins break down). And the fail-safe enduser-doable "solution" to this is simply to reboot their box. The default button in that dialog to do the reboot immediately, but you may also choose "later" (which merely exits the installer).
This behaviour is not that much different from many linux boxes, it's merely a different default button selected. When you upgrade a Debian Linux box from Etch to Lenny, the libc-upgrade also restarts a few services who are known to rely on the nss-functions of libc, but also gives this recommendation:
If other services have begun to fail mysteriously after this upgrade, it is probably necessary to restart them too. We recommend that you reboot your machine after the upgrade to avoid NSS-related troubles.
Of course, if you know what you're doing and the guts of your OS inside out, you can work around the reboot, but most MacOS X users don't do so (and often have choosen the Mac exactly for the reason that they don't have to mess around with their computer more then necessary in order to make it work).
So in case to be sure, you'll do the reboot sooner or later anyway.
Simply uninstall safari and iTunes on your server, depending on your mileage for "server", they're useless anyway and shouldn't be installed on the box.
Yes, the Apple gui installer doesn't give you a way to do so, but there are tons of free 3rd party tools to accomplish this.
From the early days of iTunes in Mac OS 9, there has been support for plugins. And in those "good old days", there were plugins written in order to directly support syncing e.g. the RIO MP3 players.
Scroll down to the last paragraph at rioaudio.com/itunes/: there's still such an old plugin to support some RIO players under OS 9 (according to the website, newer iTunes releases do support all RIO players out of the box).
As at least to my knowledge, the itunes plugin interface hasn't disappeared or cropped down, I guess it's not that a strange issue to write a plugin to support the media player of your dreams or to import music the way you'd like to (even if it's raw filescanning).
And actually, the "well structured music folder" is something that has to be done "by hand", and as such, is prone to errors or mis-interpretation. If you're taking a look at how non-geeks do work with their computer, it is quite a luck that iTunes does take that part (by default). According to my observation, very many non-geeks save all their files under non-descriptive names into a single "documents" folder on their local hard drive instead of the backed up network share, Linux-Newbies save their files right into their home directory. After about half a year, those people don't find their old documents anymore.
Remember the early days of mp3? What happens if you give such people the tools to create a dozen files by inserting an audio cd: track1-12.mp3 in some "music"-directory. Of course, there are no correct tags and those few even include spelling errors.
iTunes by default takes over that task and manages that data in an easily understandable way ("importing"). Of course, you can use OS X folder actions to automatically import mp3 files from your web browsers download folder. In my view, the download folder is merely some kind of scratch directory where new files arrive and need to be manually moved to the places they belong to.
iTunes also takes and writes metadata from and to mp3/aac tags and caches them into a local pseudo-database, as such enforcing "correct" tags. In case you're sending an itunes-managed file (e.g. from a podcast or your free indie music) to your friend, the file contains correct metadata. Your friend doesn't need to recreate the same filesystem structure on his own, she can simply drop it into the window of her media player and is done.
Compared to the early days of mp3, enforcing tags is actually a really cool feature. And auto-structuring the files isn't that bad as well, as it automatically gives you about the very same structure you're currently managing on your own.
In my metropolitan area, electronics recycling is pain. Only two locations, neither of them convenient in location of hours. The unstaffed locations do not accept things like batteries or electronics.
Well, in some countries, the situation is a little different. E.g. recycling batteries is a no-brainer in germany: every store who sells batteries is required by law to accept used batteries and return them for recycling.
Usually they do set up dropboxes, which are collected "for free" (to the shop) and properly recycled or disposed (depending on the type and state of battery) by a foundation which is paid by the battery industry. The english subsite of www.grs-batterien.de gives an overview on the subject.
Additionally to those boxes, many cities do install special public waste containers for used batteries in living areas (there are at least two containers in my street and they're located next to the glass recycling bins).
In general I would hate to see laws that required or forbade removable batteries. What I would like to see is more retailers forced to take back electronics that they sell, perhaps with a small discount if you buy an equivalent device.
There has been a move in that direction every once in a while in the EU and some countries in Europe have made their own laws to do so, but a "generic" EU regulation isn't in sight.
In Switzerland shops are forced to take back used electroncics for recycling - but quite a lot shops refuse to do so and don't remember the law.
In Austria, you can return used electronics to the manufacturer: remove any batteries from the device and recycle them seperately, drop the device in a box with a special label and send it (via postal service) back to the manufacturer for free. And usually, you can also bring back the old device to your shop for recycling.
The situation in Germany is unluckily roughly the same as yours: my city (300k people living here) has two off-site locations with limited opening hours. Better shops also accept used electronics for recycling. Its also common for mail order to offer taking care of recycling upon ordering a new replacement device, often for a low transportation fee. So if you're ordering a new fridge via mail order, you may also order a "recycling option" for 10 Euros which makes the delivery service dropping the new fridge at your place return your old fridge to a recycling center.
What's going to be more expensive: A massive NAT box or an IPv6-enabled router (as many already are)?
What's going to be more expensive: Adding NAT buster support into many apps, or using IPv6 (many apps are already IPv6-aware)?
Well, except those guys designing protocols by copying FTP, most client applications actually don't have a problem running with NAT.
Please note that many DSL and cable ISPs provide their customers with cheap "WLAN routers", who often really are plain NAT devices and so by today, you do have millions of people sitting behind NAT gateways and so NAT is a wide-spread "solution" in the industry.
NAT is in use at millions of sites, while the current worldwide amount of IPv6 traffic is merely something in the range of roughly less than a gigabit and about this well "tested" in real life. So if your router does run IPv6, it's not sure that this router will be able to run IPv6 reliably once IPv6 is widely deployed.
At the APNIC 26 conference last month, NTT presented some ballpack numbers for how many people can be comfortably put behind NAT. They're not encouraging. Basically, the common "Web 2.0"-type apps open a lot of background connections, which chews through your ephemeral port space quickly, limiting the number of people that can be NATted. Google echoed those claims loud and clear: "AJAX applications break behind excessive NAT."
Also, consider that by 2012 we'll have run out of public IPv4 addresses. But only 25% of Earth's population will be online. Do you propose to put another 3.5 billion people behind NAT? I'm pretty skeptical that NAT can handle that load.
While NAT will likely be needed in the short term to deal with IPv4 address exhaution, I'm highly skeptical of its long-term scalability.
Well, "Ajax applications will break under NAT", "tomorrow we'll run out of IPv4 addresses" and "the whole world wants to be online, so we do need more IP space" are very close to FUD.
And may I point out that a clear majority of those 3.5 billion people are much more in nead of clean, drinkable freshwater, food to eat and at least some kind of health care rather than the ability to watch sneezing pandas on youtube?
Doesn't it make more sense to keep using the bar code system they already have on passports with the data tied to the passport stored remotely ? Storing data on the passport itself makes no sense.
Storing data on the passport itself is necessary if you don't have access to that database or where database access were too slow to really use.
If you don't know what I think of "slow": just use any ATM in some foreign country. It does take about one or two minutes to verify that the data from the magnetic stripe along with the PIN are for some existing bank account which has enough money to cash via the ATM.
Usually, as a tourist you need your passport not only for entering or leaving a country but also to prove that the flight ticket belongs to you. And if you've been caught driving an mph either too fast or way too slow (because you're not yet fully accustomed of the speed limits), you also need some proof for the police that that easily-copied international driving license and the online assigned, self-printed tourist visa really are backed by "really" official document.
If it takes 10 minutes or so for the police in my own country to verify that everything is fine by checking some online database, I suspect that it would take a while longer for some police officer in some foreign country equipped with some online terminal which doesn't match the exact requirements of the needed online database (e.g. to enter some umlauts or otherwise accented characters).
And from the people I speak to, lots of people aren't visiting the US due to all the information that the US requires, and the way they're treated at Immigration.
AOL.
After becoming aware of the japanese customs demanding fingerprints from every visitor, I've canceled the thought of spending at least a few weeks of my vacation this year in Japan.
Yep, I can guess your response: Well don't come here then, we don't want you anyway.
When the customs office treats me like a criminal by taking mugshots or fingerprints and storing them along with a bunch of strictly personal data about myself, and all this just for the reason that "tourist" somehow shares a few letters with "terrorist" and the fact that the same customs office wouldn't dare to treat their own citizens the same way, the country literally shouts at me "go away".
Plus, now your machine doesn't even have to be turned off for someone to remove it to a forensic lab: introducing HotPlug. HotPlug only works in places where you could introduce an UPS to the plug connectors while they're still (electrically live) connected to the socket.
However, many other countries do use power plugs and wall sockets where you can't get access to the plug connectors while the plug is still in the socket. E.g. see CEE7/7, which is the de-facto-standard for many european countries or http://en.wikipedia.org/wiki/Domestic_AC_power_plugs_and_sockets for various plugs and sockets. Some of them might be intercepted via HotPlug, but I guess that at least for CEE7/7 or IEC connectors it might be quite hard to "HotPlug" them.
Sidenote: Bavaria is the state who (back in the 1970s, when there were only publicly operated TV stations in Germany) forbid broadcasting of the first "Sesame Street" episodes, as it also talked about "less privileged children" and the state of Bavaria stated that there were no such thing like "less privileged children" in Bavaria and this would only confuse children.
Günther Beckstein has several "controversal" opinions, not commonly shared among germans (most often even not his own political party). He wants to introduce more CCTV surveillance cameras in public areas, deport foreigners simply upon the first suspect of terrorism and has just been quoted with "'Killer-games ' should be classified like children pornography, so that there are more noticable penalties". Günther Beckstein also strongly votes for higher penalties on shoplifting and grafitti sprayers. Luckily for civil rights, most of his thoughts actually never made it to reality.
The "killer games" term references to first-person shooters and has just been a hot issue during the last few weeks, when someone has been running amok at his former school (at least, he didn't kill anybody but himself). But even that student has been depicted as much introverted, openly aggressive wearing military-style clothing, he has been much into weapons, listened to recordings and glorified the columbine high school shooter, announced his amok run on a website, built a CS-map of his school and the next day, he would've faced a charge for owning a fire weapon without a valid license. "Of course", that guy has also been playing counter strike and so there have been much politicians into thinking that playing FPS games is a clear sign for such behaviour, although many psychologists have stated that "playing FPS" is merely a side symptom and many gamer's communities have said that most of their game play focusses on team work and communication, so it is rather unusual symptom for an experienced player.
From as far as I've heard of, the current ideas how to implement that EU directive in Germany will be more in the way...
The ISP has to store the data for exactly 6 months on his own cost at his own site. No remotelogging to some agency, no costsharing.
This does apply to "communication" logs - httpd, file transfer, mail and dialups. Some sites (web hosts, freemail services,...) are logging way more than 100 Byte a day, and so the cost for that logging isn't that low as some readers might guess. It doesn't mean that any ISP has to tcpdump or mirror their traffic, but any of their own server who is logging something would have to comply to this rule.
The ISP may not use that data for his own purpose other than as much as needed by the user's contract (e.g. billing, statistics) or internal debugging. If they're looking up an IP adress for net abuse without any legal document asking them to do so, that evidence is to be rejected in any court and the one presenting that "evidence" is likely to get into some legal trouble.
At least from some legal point of view, this restricts the use to comply with both privacy as well as the EU snooping directive.
This single court decision is binding (when not calling for appeal) and D-Link in between acknowledged that they won't challenge the decision.
It's important to note that you can't quote a precedent in Germany, as the german law doesn't take care of previous cases. E.g. a court in munich decided back in 2004, that the GPL does apply (back then: for netfilter/iptables) and can be enforced (in that case, against Sitecom, another router vendor).
You can present that there have been a dozen decisions in one direction, but usually that doesn't have any impact on the court's decision. So this single court decision can't be regarded as a test case, but it still improves the situation for defending the GPL a little bit - as a reference, but not as a precedent. In fact, there are quite often cases where different highest regional courts do state completely opposite decisions for very much comparable, but arguable cases. For example, german websites have to show an imprint, stating a few legal things along with contact information (full name, postal address, usually phone number and email address). OLG Hamm decided that you don't have to state your phone number, an email adress with an often checked account ist satisfying, while OLG Cologne previously stated a very different view in a much noted court decision.
This recent D-Link case is still somehow cumbersome:
D-Link tried to rely on section 2 of the GPL, which might be unenforcable under german law or in some interpretation violate german or even european law. As the GPL doesn't contain a severability clause (it doesn't render a whole document invalid just because a part of it is regarded as invalid), this might've taken down the whole GPL (this needs to be considered for new versions of the GPL!). D-Link argued that without a valid license, this open source software would've been completely free from any rights or licenses.
The court revoked that idea, because setting some code under GPL still doesn't remove your copyright on that code; so D-Link could've been challenged for copyright infringement and all parties settled for section 4 of the GPL, completely ignoring anything out of that scope.
The court's decision (of course in german) can be found at http://www.jbb.de/urteil_lg_frankfurt_gpl.pdf (scanned fax pages). The name "D-Link" has been striked out, due to some other law thingie (jbb.de are the lawyers for gpl-violations.org any may not offend privacy rights of the other party), but the rest is still complete.
who would seriously take their ipod into a combat situation?
In a documentation video recently aired on Arte, they've tried to take a look at the situation of "normal" people, US soldiers in Iraq along with deserters who went to Canada. One soldier mentioned that the radios in recent tanks have an additional line-in input hidden on the back, where they used to plug in portable cd players . Incoming radio messages automatically turn down that devices volume, so it's "perfect" for listening some music while waiting for something to happen. A deserter being asked on that subject stated that some teams even modified their field radios in a similar way in order to listen to some "encouraging" music while actually firing... of course, it is not allowed to both modify the equipment or even let oneself disattract from the situation by listening to some music, but those guys were pretty serious on that thing.
It sounds silly and socially non-acceptable, but what exactly prevents people from plugging in their iPod (or other MP3 player) instead of some dated discman into such a radio?
Come on. Getting the reason for some Linux kernel panic is the same story.
Faulty hardware like defective RAM or even a broken CPU fan running at only half the speed can be the reason, software errors in the base system or the kernel can be the reason as well as any "third-party" software (who sometimes also do install some kludgy kernel extension).
Yesterday, my (PC) workstation refused to boot - no beep, no picture, nothing except the power led. It turned out that the fan on the graphics adapter, who went into smoke a few weeks ago. Obviously, the other fans in the case as well as my low usage of GPU-intensive work kept that adapter alive, just until I had to turn off the machine yesterday.
Years ago, I symlinked "SEGFAULT/signal 11" to "broken memory". Nowadays, I don't do so any longer. It may be some fine line on the main board, some defective CPU fan,... only if it comes to hardware. And after I've seen a valgrind session running a simple "php.cgi -i" (php_info()), I ignore any PHP if it is throwing those segfaults.
Back to the panic screen: of course, you can present a dialogue stating that something went wrong, and it may be hard- or software, stating various things which "might" be the issue, but in the end, close to everything is possible. But what does the user get out of this other than "there's something broken and we're forcing you to reboot anyway"? If those messages do appear a dozen times a day, even the most dumb user would ask some engineer to take a look at it - and those guys will do check/var/log/ for suspicious entries.
Slashdot Mirror
Geoff Huston wrote an article about comparing 6to4 to IPv4 in terms of failure rate - and found out that about 0.2% of IPv4 connections to his web server were also broken. Geoff's article also provides insight why exactly a large percentage of 6to4 connections to his web server failed: routing packets around the planet just because a lack of 6to4 gateways and in three out of four issues, some broken firewall dropping 6to4 packets.
Issues A and B also doesn't necessarily mean that IPv4 is permanently broken but "occasionally".
For example, every mobile carrier in my country deploys Large Scale NAT/Carrier Grade NAT for IPv4, but in order to max out those boxes, they're running with very low session timeout settings. Their NAT routers silently drop my session when a tcp connection is idle for longer than a few seconds. While web browsing "usually" works, things like IMAP sessions very often do break and reconnect, for interactive use I'm forced to run ssh-sessions with "ServerAliveInterval 7". One of those carriers even temporarily blocks access to Apple's iTunes store - maybe because the iTunes store is known to eat up to 300 parallel NAT sessions for a single user (compared to roughly 20-30 for "usual" web surfing). When accessing some very slow web server, the NAT session timeout also kicks in, resulting in my browser "endlessly" loading the same page.
Right now, the same carriers don't yet offer IPv6, so technically, they're forcing me to issue "A". Once they do offer IPv6, my mobile internet access is likely to be issue "B".
The same experiment can actually work out very differently. At Google's IPv6 implementors conference in summer 2010, a japanese ISP reflected about the very :-)
same experiment like heise.de or the World IPv6 day do of adding AAAA-records for a day.
They've been doing IPv6 for years now, including hosting via IPv6. When they added AAAA-records for their very large japanese portal site biglobe.ne.jp, they lost about 5% of page views immediately and 5 minutes later, their phone started ringing endlessly. A few hours later, they've chosen to cancel the experiment by removing AAAA-records from their DNS.
In my mind, many japanese ISPs have been using and offering IPv6 access for years now, but there haven't been any major services available via IPv6 in Japan, so the actual IPv6 traffic has been very low and most people weren't aware that their IPv6 setup is simply broken. Maybe even Yahoo's and Google's often-quoted "0.025% of users do have IPv6 issues" bases on Japan being largely broken in terms of IPv6 service while the rest of the world may run IPv6 without any issues
Well, Germany is quite a very different issue. Most large german access (DSL/broadband/dialup) ISPs don't yet support IPv6 and the de-facto standard-dsl-router range of most ISPs (AVM's Fr!tz-box) didn't support any kind of IPv6 at all until quite recently. Even now, IPv6 is something hidden deep in their menues and actually needs to be manually turned on. German web hosting consists of a few large companies, where support of IPv6 is currently left as a DIY-option for dedicated servers and not for any shared hosting plans. On the other hand, close to every ISP peers via IPv6, is running 6to4 gateways and happily runs IPv6 on their own networks, but IPv6 isn't yet used for any actual major public service, so in theory, IPv6 shouldn't be that hard to get working in Germany today ... but for today, IPv6 in Germany is actually VERY poor.
To illustrate how worse IPv6 in germany is, check the TLD stats at Hurricane Electric, compare the amount of AAAA-records vs. the amount of A-records. .net, .org, ...), there's one AAAA-record for roughly about every 90 A-Records. For .de, only about one out of thousand A-Records do have an AAAA-record. That's a ten-fold in being worse!
For about every TLD (.com,
So heise.de didn't really venture a lot when they turned on IPv6, as even far less users in Germany actually do use IPv6 than in any other country. However, they've still done something very intelligent: once German Internet access ISPs do turn on IPv6 connectivity for their customers and customers notice about heise.de being unreachable, heise.de users are already aware that heise.de has been served via IPv6 for months without any problems, so any brokenness must be related to their own ISP (or their personal setup). They'll directly complain to their ISP and won't blame heise.de.
In fact, there are quite a few people out there using Anycast for TCP-sessions. It's really a matter on what timescale you're looking at. The networking guys see TCP as something to use for long-living connections - e.g. a BGP session running for days, weeks or even months. A flapping route in this setup will result in a broken session. But: what does this really mean to you? If your CDN distributes downloads which are "done" within a few minutes, such a rarely flapping route will result in a few broken sessions once a day out of millions of downloads successfully served. Compared to issues like non-working DNS, overloaded servers and filled lines, that's nothing and can actually enhance the overall CDN service.
A nice paper to read is this one from Matt Levine. He's working for a CDN provider using TCP-Anycast for years now and sums up the most important issues on TCP-Anycast.
Basically the most important one is that your anycasted servers really have to be spread far enough so that flapping routes at some peering point won't matter. As a rule of thumb, put one CDN loadbalancer on the US east coast, one to the US west coast, another one to western europe, one to Australia and one in Hong Kong. If you'd like to put multiple CDN loadbalancers to one continent, leave space between them, e.g. one box for each country/state.
VG Musikedition (the other "club" represented by GEMA), sums up this issue and
VG Musikedition on photocopies in Kindergarten outlines their "new" offer. You may want to give e.g. Google Translate a try if you don't understand the german language.
Until recently, Kindergartens weren't permitted at all to copy single song sheets, their only option were to buy the books of those song sheets, but those books again didn't permit copies at all. However, Copyright expires 75 years after death of the writer; and those songs may be freely copied.
Now, there's also the option for Kindergartens to pay a fee of 56 Euros (plus 7% VAT) per year (€44,80 for Kindergartens operated by churches or cities) which permits up to 500 copies of song sheets. For example, for 112 Euros (plus 7% VAT), up to 1000 copies are permitted.
The odd thing with this option is, that kindergarten teachers then do have to keep an account on the amount of copies being taken for songs whose writer is dead for at least 75 years: bureaucracy at its best.
Kindergarten age kids in Germany can read sheet music? I'm impressed...
The sheets are not for the children but their parents, so they may sing songs which have been taught their children.
Okay, without sarcasm.
Back in the "good ole days of the internet", IP-addresses were given out as permanent property and there's about no legal way for IANA or the current RIRs to recall those IP-adresses.
Nowadays, IP-addresses are given out as some kind of semi-permanent lease. For example, your RIR may offer you a larger allocation than a requested one, but they also may require you to hand back your old allocation after a few months to allow renumbering your old IP space.
Even forcing those organizations to hand out their /8 does give us about two more years, that's simply ridiculous. IPv6 has been in work since last century's nineties, experimental networks like 6bone were closed in 2006 after IPv6 has been declared as being stable for "production use".
Even older operating systems like Linux 2.4 and Windows 2003 have IPv6-stacks officially being seen as "stable for production use" (Windows 2003 is lacking IKE for IPsec, but that's all about it). Whoever isn't capable of deploying IPv6 within the next two years deserves being doomed.
You forgot to mention a few other issues.
IANA reserved 224.0.0.0/4 for "multicast"-usage, that's the equivalent of 16 /8-sized networks.
Renumber multicast!
We also need to replace RFC 1918's wasteful use of 10.0.0.0/8. No organization ever needs 16 Million IP addresses, even Google has a fraction of physical servers than this.
While I'm thinking about it: RFC 3330 spends more than 16 Million IP addresses for a single box. ::1/128 and return 127.0.0.0/8!
Whoever is using this 127.0.0.0/8: please do renumber to e.g.
A problem that utterly destroyed the work of amateurs like DaVinci, Michaelangelo, and Raphael, right?
During his last two decades, Ludwig van Beethoven lost his hearing. He was completely deaf when he composed
his ninth symphony (famous for e.g. "Ode to Joy"). That doesn't mean that a hearing impairment enhances songwriter skills.
Imagine the works of those artists if they weren't bound to cover the area they're working on.
They might've raised the bar of perceived perfection to even higher levels.
Okay, let's recap the situation.
You said you've been using your 1&1 webspace for hosting a counterstrike mirror back in Germany during CeBIT 2000.
CeBIT 2000 took place in the end of february, so lets think about spring of 2000 back in Germany.
Most users were using simple modems, connecting to their ISP at something around 40-52 kbit downstream (depending on line quality). So for most users, it took around 2-3 minutes per MB of download.
Software back then wasn't really built to withstand any higher usage scenarios or serious DoS attacks.
Now imagine what might've happened. Got it? No?
You've been hosting some file in a country where the average user would need a few hours to retrieve that file from some service which has been built for about the exact opposite of what you're doing. Once just a few people try to get your "mirror" file, they'll bring the web server down, swapping and crawling on its knees.
Your website is hosted on the same ip address than hundreds of other customers, so your website can't be moved to a different server. About the only thing the admin can do in that situation is to get those hundreds of users other than you back online is either to swap the core component of web hosting software (Apache) by "something yet to be invented", change from a non-clustered hosting service to a large-scaled cluster or simply prevent users from downloading your file.
How much crime does a better passport stop, anyway?
It doesn't prevent any crimes and was never made to do so. A passport is just a commonly accepted item for identification.
In fact, many countries don't issue a passport unless you're applying for one, and the main reason for having a passport is the point that some country other than your own would like to verify wether the one who wants to pass their borders is the one claiming to do so while making sure that a blacklist of "known bad people" won't enter their country.
If you're reading the technical specs on the passport documents, you'll note that the biometric information (JPEG of photo, hashes of fingerprint marks) is stored into an RFID chip on the actual passport, and not somewhere else.
Even just the thought of inter-connecting the millions of passport-checking locations and granting those passport-checking devices (who are under control of hundreds of different countries with tens of thousands of IT operation teams) international access to distributed giant databases is ridiculous in the eyes of anyone who ever tried to set up a mashed VPN network.
In clear: there is no online comparison or verification of those biometric data against the data provided by the passport-issueing country.
What actually may be verified is that a scanned image of the real photo matches the JPEG stored on the RFID chip and that the JPEG is cryptographically signed with a known-good signature.
And that's the point: the biometric data in an RFID chip are being used to aid against illegal passport duplication.
So in essence, the guys make copying a passport much harder.
It does no longer only take some better colour printer, sealing those printouts in plastic and wrapping it between cardboard sheets, now it also needs someone who can break public key cryptography. That's all.
The 64-bit NTP timestamp spans 136 years with a resolution of 232 picoseconds, the 128-bit NTP timestamp spans 584 billion years with a resolution of .05 attoseconds - so right from those points, NTP is good enough for your applications.
What's still problematic is a problem that NTP also tries to compensate: the network latency.
When you're receiving just two packets with exactly the same latency, you can't be sure that the third packet will be there with the same latency, so you're having an possible error rate of 33%. However, if you've seen a million packets with the same latency, your possible error rate is very close to zero, and that's why NTP can sort out the network latency problem only over time.
IIRC the law here in Germany just changed such that the onus fell on the employer to show a reason why *not* to let you work part-time.
Your're talking about an act issued in 2000, the
Gesetz über Teilzeitarbeit und befristete Arbeitsverträge (Teilzeit- und Befristungsgesetz - TzBfG) (roughly translates to "act on part-time work and time-limited job contracts").
As forcing legalese through babelfish usually gives quite "strange" results, here's my abstract in short and layman's terms:
After 6 months of being employed in a company, you may ask your company 3 months in advance to a date set by your own that you'd like to cut down your worktime to some fixed amount (less than current). If company can't show any clear, legitimate reasons (e.g. extra-ordinary costs or security) why your job can't be done part-time one months in advance to your set date, you'll be working part-time.
An extra-bonus point applies for the employee, if the company has more than 15 "regular" workers: the act expects that such "large" companies are able to compensate part-time work.
Various non-discrimination rules do apply: your net income for e.g. a 20h/week job may not be less than 50% of the net income for the same week running at 40h/week. And once there are new fulltime job offers, those offers do have to be made available to current employes, where giving those jobs to part-time employees has to be preferred. Once you're working part-time, you may also apply for more work hours.
Well, the reasoning is simply that the update swapped some shared libraries, which are in use by more than a few known applications.
Of course, Apple might say "Please restart at least your browser after updating and any other open application which somehow misbehaves", but many people do fail to do so ("well, I'll do it later" - and suddenly remember that thought once their browser plugins break down). And the fail-safe enduser-doable "solution" to this is simply to reboot their box. The default button in that dialog to do the reboot immediately, but you may also choose "later" (which merely exits the installer).
This behaviour is not that much different from many linux boxes, it's merely a different default button selected. When you upgrade a Debian Linux box from Etch to Lenny, the libc-upgrade also restarts a few services who are known to rely on the nss-functions of libc, but also gives this recommendation:
If other services have begun to fail mysteriously after this upgrade, it is probably necessary to restart them too. We recommend that you reboot your machine after the upgrade to avoid NSS-related troubles.
Of course, if you know what you're doing and the guts of your OS inside out, you can work around the reboot, but most MacOS X users don't do so (and often have choosen the Mac exactly for the reason that they don't have to mess around with their computer more then necessary in order to make it work).
So in case to be sure, you'll do the reboot sooner or later anyway.
Simply uninstall safari and iTunes on your server, depending on your mileage for "server", they're useless anyway and shouldn't be installed on the box.
Yes, the Apple gui installer doesn't give you a way to do so, but there are tons of free 3rd party tools to accomplish this.
From the early days of iTunes in Mac OS 9, there has been support for plugins. And in those "good old days", there were plugins written in order to directly support syncing e.g. the RIO MP3 players.
Scroll down to the last paragraph at rioaudio.com/itunes/: there's still such an old plugin to support some RIO players under OS 9 (according to the website, newer iTunes releases do support all RIO players out of the box).
As at least to my knowledge, the itunes plugin interface hasn't disappeared or cropped down, I guess it's not that a strange issue to write a plugin to support the media player of your dreams or to import music the way you'd like to (even if it's raw filescanning).
And actually, the "well structured music folder" is something that has to be done "by hand", and as such, is prone to errors or mis-interpretation. If you're taking a look at how non-geeks do work with their computer, it is quite a luck that iTunes does take that part (by default). According to my observation, very many non-geeks save all their files under non-descriptive names into a single "documents" folder on their local hard drive instead of the backed up network share, Linux-Newbies save their files right into their home directory. After about half a year, those people don't find their old documents anymore.
Remember the early days of mp3?
What happens if you give such people the tools to create a dozen files by inserting an audio cd: track1-12.mp3 in some "music"-directory. Of course, there are no correct tags and those few even include spelling errors.
iTunes by default takes over that task and manages that data in an easily understandable way ("importing"). Of course, you can use OS X folder actions to automatically import mp3 files from your web browsers download folder. In my view, the download folder is merely some kind of scratch directory where new files arrive and need to be manually moved to the places they belong to.
iTunes also takes and writes metadata from and to mp3/aac tags and caches them into a local pseudo-database, as such enforcing "correct" tags. In case you're sending an itunes-managed file (e.g. from a podcast or your free indie music) to your friend, the file contains correct metadata. Your friend doesn't need to recreate the same filesystem structure on his own, she can simply drop it into the window of her media player and is done.
Compared to the early days of mp3, enforcing tags is actually a really cool feature. And auto-structuring the files isn't that bad as well, as it automatically gives you about the very same structure you're currently managing on your own.
In my metropolitan area, electronics recycling is pain. Only two locations, neither of them convenient in location of hours. The unstaffed locations do not accept things like batteries or electronics.
Well, in some countries, the situation is a little different.
E.g. recycling batteries is a no-brainer in germany: every store who sells batteries is required by law to accept used batteries and return them for recycling.
Usually they do set up dropboxes, which are collected "for free" (to the shop) and properly recycled or disposed (depending on the type and state of battery) by a foundation which is paid by the battery industry. The english subsite of www.grs-batterien.de gives an overview on the subject.
Additionally to those boxes, many cities do install special public waste containers for used batteries in living areas (there are at least two containers in my street and they're located next to the glass recycling bins).
In general I would hate to see laws that required or forbade removable batteries. What I would like to see is more retailers forced to take back electronics that they sell, perhaps with a small discount if you buy an equivalent device.
There has been a move in that direction every once in a while in the EU and some countries in Europe have made their own laws to do so, but a "generic" EU regulation isn't in sight.
In Switzerland shops are forced to take back used electroncics for recycling - but quite a lot shops refuse to do so and don't remember the law.
In Austria, you can return used electronics to the manufacturer: remove any batteries from the device and recycle them seperately, drop the device in a box with a special label and send it (via postal service) back to the manufacturer for free.
And usually, you can also bring back the old device to your shop for recycling.
The situation in Germany is unluckily roughly the same as yours: my city (300k people living here) has two off-site locations with limited opening hours. Better shops also accept used electronics for recycling. Its also common for mail order to offer taking care of recycling upon ordering a new replacement device, often for a low transportation fee. So if you're ordering a new fridge via mail order, you may also order a "recycling option" for 10 Euros which makes the delivery service dropping the new fridge at your place return your old fridge to a recycling center.
What's going to be more expensive: A massive NAT box or an IPv6-enabled router (as many already are)?
What's going to be more expensive: Adding NAT buster support into many apps, or using IPv6 (many apps are already IPv6-aware)?
Well, except those guys designing protocols by copying FTP, most client applications actually don't have a problem running with NAT.
Please note that many DSL and cable ISPs provide their customers with cheap "WLAN routers", who often really are plain NAT devices and so by today, you do have millions of people sitting behind NAT gateways and so NAT is a wide-spread "solution" in the industry.
NAT is in use at millions of sites, while the current worldwide amount of IPv6 traffic is merely something in the range of roughly less than a gigabit and about this well "tested" in real life. So if your router does run IPv6, it's not sure that this router will be able to run IPv6 reliably once IPv6 is widely deployed.
At the APNIC 26 conference last month, NTT presented some ballpack numbers for how many people can be comfortably put behind NAT. They're not encouraging. Basically, the common "Web 2.0"-type apps open a lot of background connections, which chews through your ephemeral port space quickly, limiting the number of people that can be NATted. Google echoed those claims loud and clear: "AJAX applications break behind excessive NAT."
Also, consider that by 2012 we'll have run out of public IPv4 addresses. But only 25% of Earth's population will be online. Do you propose to put another 3.5 billion people behind NAT? I'm pretty skeptical that NAT can handle that load.
While NAT will likely be needed in the short term to deal with IPv4 address exhaution, I'm highly skeptical of its long-term scalability.
Well, "Ajax applications will break under NAT", "tomorrow we'll run out of IPv4 addresses" and "the whole world wants to be online, so we do need more IP space" are very close to FUD.
And may I point out that a clear majority of those 3.5 billion people are much more in nead of clean, drinkable freshwater, food to eat and at least some kind of health care rather than the ability to watch sneezing pandas on youtube?
Doesn't it make more sense to keep using the bar code system they already have on passports with the data tied to the passport stored remotely ? Storing data on the passport itself makes no sense.
Storing data on the passport itself is necessary if you don't have access to that database or where database access were too slow to really use.
If you don't know what I think of "slow": just use any ATM in some foreign country. It does take about one or two minutes to verify that the data from the magnetic stripe along with the PIN are for some existing bank account which has enough money to cash via the ATM.
Usually, as a tourist you need your passport not only for entering or leaving a country but also to prove that the flight ticket belongs to you. And if you've been caught driving an mph either too fast or way too slow (because you're not yet fully accustomed of the speed limits), you also need some proof for the police that that easily-copied international driving license and the online assigned, self-printed tourist visa really are backed by "really" official document.
If it takes 10 minutes or so for the police in my own country to verify that everything is fine by checking some online database, I suspect that it would take a while longer for some police officer in some foreign country equipped with some online terminal which doesn't match the exact requirements of the needed online database (e.g. to enter some umlauts or otherwise accented characters).
And from the people I speak to, lots of people aren't visiting the US due to all the information that the US requires, and the way they're treated at Immigration.
AOL.
After becoming aware of the japanese customs demanding fingerprints from every visitor, I've canceled the thought of spending at least a few weeks of my vacation this year in Japan.
Yep, I can guess your response: Well don't come here then, we don't want you anyway.
When the customs office treats me like a criminal by taking mugshots or fingerprints and storing them along with a bunch of strictly personal data about myself, and all this just for the reason that "tourist" somehow shares a few letters with "terrorist" and the fact that the same customs office wouldn't dare to treat their own citizens the same way, the country literally shouts at me "go away".
I take those words for granted.
However, many other countries do use power plugs and wall sockets where you can't get access to the plug connectors while the plug is
still in the socket. E.g. see CEE7/7, which is the de-facto-standard for many european countries or
http://en.wikipedia.org/wiki/Domestic_AC_power_plugs_and_sockets for various plugs and sockets. Some of them might be intercepted via HotPlug, but
I guess that at least for CEE7/7 or IEC connectors it might be quite hard to "HotPlug" them.
Sidenote: Bavaria is the state who (back in the 1970s, when there were only publicly operated TV stations in Germany) forbid broadcasting of the first "Sesame Street" episodes, as it also talked about "less privileged children" and the state of Bavaria stated that there were no such thing like "less privileged children" in Bavaria and this would only confuse children.
Günther Beckstein has several "controversal" opinions, not commonly shared among germans (most often even not his own political party). He wants to introduce more CCTV surveillance cameras in public areas, deport foreigners simply upon the first suspect of terrorism and has just been quoted with "'Killer-games ' should be classified like children pornography, so that there are more noticable penalties". Günther Beckstein also strongly votes for higher penalties on shoplifting and grafitti sprayers. Luckily for civil rights, most of his thoughts actually never made it to reality.
The "killer games" term references to first-person shooters and has just been a hot issue during the last few weeks, when someone has been running amok at his former school (at least, he didn't kill anybody but himself). But even that student has been depicted as much introverted, openly aggressive wearing military-style clothing, he has been much into weapons, listened to recordings and glorified the columbine high school shooter, announced his amok run on a website, built a CS-map of his school and the next day, he would've faced a charge for owning a fire weapon without a valid license.
"Of course", that guy has also been playing counter strike and so there have been much politicians into thinking that playing FPS games is a clear sign for such behaviour, although many psychologists have stated that "playing FPS" is merely a side symptom and many gamer's communities have said that most of their game play focusses on team work and communication, so it is rather unusual symptom for an experienced player.
If they're looking up an IP adress for net abuse without any legal document asking them to do so, that evidence is to be rejected in any court and the one presenting that "evidence" is likely to get into some legal trouble.
At least from some legal point of view, this restricts the use to comply with both privacy as well as the EU snooping directive.
ista
This single court decision is binding (when not calling for appeal) and D-Link in between acknowledged that they won't challenge the decision.
It's important to note that you can't quote a precedent in Germany, as the german law doesn't take care of previous cases.
E.g. a court in munich decided back in 2004, that the GPL does apply (back then: for netfilter/iptables) and can be enforced (in that case, against Sitecom, another router vendor).
You can present that there have been a dozen decisions in one direction, but usually that doesn't have any impact on the court's decision. So this single court decision can't be regarded as a test case, but it still improves the situation for defending the GPL a little bit - as a reference, but not as a precedent. In fact, there are quite often cases where different highest regional courts do state completely opposite decisions for very much comparable, but arguable cases.
For example, german websites have to show an imprint, stating a few legal things along with contact information (full name, postal address, usually phone number and email address). OLG Hamm decided that you don't have to state your phone number, an email adress with an often checked account ist satisfying, while OLG Cologne previously stated a very different view in a much noted court decision.
This recent D-Link case is still somehow cumbersome:
D-Link tried to rely on section 2 of the GPL, which might be unenforcable under german law or in some interpretation violate german or even european law. As the GPL doesn't contain a severability clause (it doesn't render a whole document invalid just because a part of it is regarded as invalid), this might've taken down the whole GPL (this needs to be considered for new versions of the GPL!). D-Link argued that without a valid license, this open source software would've been completely free from any rights or licenses.
The court revoked that idea, because setting some code under GPL still doesn't remove your copyright on that code; so D-Link could've been challenged for copyright infringement and all parties settled for section 4 of the GPL, completely ignoring anything out of that scope.
The court's decision (of course in german) can be found at http://www.jbb.de/urteil_lg_frankfurt_gpl.pdf (scanned fax pages).
The name "D-Link" has been striked out, due to some other law thingie (jbb.de are the lawyers for gpl-violations.org any may not offend privacy rights of the other party), but the rest is still complete.
In a documentation video recently aired on Arte, they've tried to take a look at the situation of "normal" people, US soldiers in Iraq along with deserters who went to Canada. One soldier mentioned that the radios in recent tanks have an additional line-in input hidden on the back, where they used to plug in portable cd players . Incoming radio messages automatically turn down that devices volume, so it's "perfect" for listening some music while waiting for something to happen.
A deserter being asked on that subject stated that some teams even modified their field radios in a similar way in order to listen to some "encouraging" music while actually firing
It sounds silly and socially non-acceptable, but what exactly prevents people from plugging in their iPod (or other MP3 player) instead of some dated discman into such a radio?
ista
Come on. Getting the reason for some Linux kernel panic is the same story.
... only if it comes to hardware. And after I've seen a valgrind session running a simple "php.cgi -i" (php_info()), I ignore any PHP if it is throwing those segfaults.
/var/log/ for suspicious entries.
Faulty hardware like defective RAM or even a broken CPU fan running at only half the speed can be the reason,
software errors in the base system or the kernel can be the reason as well as any "third-party" software (who sometimes
also do install some kludgy kernel extension).
Yesterday, my (PC) workstation refused to boot - no beep, no picture, nothing except the power led. It turned out that the fan on the graphics adapter, who went into smoke a few weeks ago. Obviously, the other fans in the case as well as my low usage of GPU-intensive work kept that adapter alive, just until I had to turn off the machine yesterday.
Years ago, I symlinked "SEGFAULT/signal 11" to "broken memory". Nowadays, I don't do so any longer. It may be some fine line on the main board, some defective CPU fan,
Back to the panic screen: of course, you can present a dialogue stating that something went wrong, and it may be hard- or software, stating various things which "might" be the issue, but in the end, close to everything is possible. But what does the user get out of this other than "there's something broken and we're forcing you to reboot anyway"? If those messages do appear a dozen times a day, even the most dumb user would ask some engineer to take a look at it - and those guys will do check