I know now - I recently read that you can easily change from Knoppix to debian. I should have known that in September! Why don't debian just use Knoppix as the installer in the first place?
SuSE don't have ISOs to download, but the far better option anyway is just to download the boot&initial install ISO (quite small) and then have an FTP install. This way SuSE would only download the packages you are actually installing. -> Much faster and less hassle burning.
I couldn't get Debian stable to run either - the problem was the RAID controller on the motherboard of my new box. So (stupidly, I agree) I downloaded and burned the 7 CDs of a "testing" snapshot just to find out that the installer was still buggy and at some point very early on in the installation process complained that it couldn't find the files it was looking for and aborted the install... And I couldn't find a way around that. It worked flawlessly with Mandrake 9.2 (I think due to a newer kernel that supported my RAID driver).
It should preserve user data and backups of
old system configuration, though. The new system is a clean Debian system,
it's not a chroot, and no traces from the old distro should keep around,
but backups.
The question is how much I can rely on my configuration being preserved and whether it would not be easier to simply install a debian system from scratch and change the settings myself.
Welcome to the club - that was exactly my mistake (maybe 3 or 4 years ago). The spam started showing up a day after I posted (only a couple a day initially) - now I am at ~70 spam e-mails every day.
Yes (although the freeze only seems temporary for me), but I always find it annoying anyway when PDFs are loaded in browser windows/tabs. What's the point? Much better to have it load in Acrobat Reader separately as all menus are then available. Set under Tools -> Options -> Downloads -> Plugins (disable PDF plugin). Then, when you click on a PDF link the next time it will ask you what to do, and you can simply choose to always automatically open PDF files with Acrobat Reader.
You are correct. I hadn't thought of the port remapping. How do NAT boxes usually remap ports? Could you infer something from the packets sent to the negotiating server? IE, if the remapped ports are always increased by one, you could try and guess the next port that will be sent from? And if you have a good guess a couple of times (always trying with a handful of ports), there might be a good chance that a connection can be established. Sounds a bit dodgy but it could work...
Yes, I had these problems with the invalid file list as well. They disappeared after choosing a different mirror. Don't know why though. Configure your mirrors for urpmi here: http://plf.zarb.org/~nanardon/
I guess you could have the third party machine get the fiewall source port of A (if you send your opening packet to it instead) and pass that on to B, but wouldn't most NAT firewall writers check to make sure that all mapped ports are remote address specific? That is, if 23.34.55.66 is where my host A sent its original UDP packet to, I'm not going to accept responses from some other IP address.
Well, you use a third-party server to find out the IP address and local port of the other side - and then you just start sending UDP packages. The first few will be lost, but then it should work.
I guess you would have to swap a few more things than IP and port numbers if there is a proper firewall in the way, but I believe that it should be possible - I am not sure if it would work with a standard TCP/IP stack though as you would have to send a SYN/ACK withouth having received a SYN. What would need to happen is the following. You have two computers behind NAT that want to speak to each other, A and B. You have an "enabling" public server C. A and B exchange IP addresses/ports and the ID of the SYN packet you are going to send through C and then agree on a time (i.e. in 0.1s) when you would send SYNs and SYN/ACKs. This should trick the firewall and NAT router into thinking that the SYN/ACKs are the replies to the SYNs sent earlier and open a connection. The problem could be that a firewall could drop outgoing SYN/ACKs without having passed through a SYN before. It might be easier with UDP (which would be good enough for VoIP)? Disclaimer: I don't really know the details here - are there any TCP/IP and firewall wizards around who could see whether this would be possible?
The problem with this theory is that it is utter nonsense.
It is not a theory. It is, as you correctly state, a hypothesis. And I for one think there is a lot of merit in stating such a hypothesis. It would be strange if stockmarket prices would not reflect the valuation (given informational constraints) of the company.
In the SCO cases, I think besides a payoff from law suit and licence sales investors/speculators are betting on a buyout by someone (not IBM though probably). That's why I don't go short (too much risk), because a buyout is possible while a win of the law suit is very improbable.
These are the same dimbulbs who gave us the tulip mania a century ago ($3000.00 for one bulb?!?)
It was actually in the early 17th century, and if I remember correctly, the price for the most expensive bulbs was (in today's money) easily a million dollars (per bulb!).
The relevant organisation is the EEA (European Economic Area) and not the EFTA. EEA = EU + Norway, Iceland and Liechtenstein. EFTA = EEA + Switzerland - EU. Switzerland does not form part of the Single Market, while Norway, Iceland and Liechtenstein do. (I believe they retain the right to veto European Directives, so that they don't apply in their respective countries.)
These are from the low-res black and white hazard avoidance camera
In fact, all but the bottom ones are from the Navigation cameras, which sit on top of the mast as well, just inside of the Hi-Res Panorama Camera. The Nav Cam has a resolution of 512x512, but these pictures were taken with 256x256. The Panorama Cam has a resolution of 1024x1024.
The information in the article is incorrect. 24 Mbit (or 3 MBytes) of data were received in total over a couple of minutes at a speed of 128 kbps (This was for the first pass of the Odyssey orbiter). 128 kbps relates to the transmission speed from the orbiters to the Deep Space Network on Earth. That's not broadband, but it's not bad either, given the distance! The direct link from the rover to Earth will be much slower with around 12 kbps (but longer time periods of communication can be achieved each day compared to passes of the orbiters). However, this will only happen once the high-gain antenna is operable (which will probably be later today) and the Earth is visible from the rover.
Slashdot Mirror
I doubt that a *business* is protected by the *consumer* protection regulations.
I know now - I recently read that you can easily change from Knoppix to debian. I should have known that in September! Why don't debian just use Knoppix as the installer in the first place?
SuSE don't have ISOs to download, but the far better option anyway is just to download the boot&initial install ISO (quite small) and then have an FTP install. This way SuSE would only download the packages you are actually installing. -> Much faster and less hassle burning.
Just curious, why would you go back to SuSE 8.0 and not use SuSE 9.0 (That's been available on the FTP server for a while).
I couldn't get Debian stable to run either - the problem was the RAID controller on the motherboard of my new box. So (stupidly, I agree) I downloaded and burned the 7 CDs of a "testing" snapshot just to find out that the installer was still buggy and at some point very early on in the installation process complained that it couldn't find the files it was looking for and aborted the install... And I couldn't find a way around that. It worked flawlessly with Mandrake 9.2 (I think due to a newer kernel that supported my RAID driver).
Welcome to the club - that was exactly my mistake (maybe 3 or 4 years ago). The spam started showing up a day after I posted (only a couple a day initially) - now I am at ~70 spam e-mails every day.
Yes (although the freeze only seems temporary for me), but I always find it annoying anyway when PDFs are loaded in browser windows/tabs. What's the point? Much better to have it load in Acrobat Reader separately as all menus are then available. Set under Tools -> Options -> Downloads -> Plugins (disable PDF plugin). Then, when you click on a PDF link the next time it will ask you what to do, and you can simply choose to always automatically open PDF files with Acrobat Reader.
This NEC weighs 61 kgs. So not really a feather :-)
What is so special about this? Is it the size? Is this the first 61-inch screen? Honestly - I don't know!
You are correct. I hadn't thought of the port remapping. How do NAT boxes usually remap ports? Could you infer something from the packets sent to the negotiating server? IE, if the remapped ports are always increased by one, you could try and guess the next port that will be sent from? And if you have a good guess a couple of times (always trying with a handful of ports), there might be a good chance that a connection can be established. Sounds a bit dodgy but it could work...
Yes, I had these problems with the invalid file list as well. They disappeared after choosing a different mirror. Don't know why though. Configure your mirrors for urpmi here: http://plf.zarb.org/~nanardon/
It all seems pretty dark to me on the pictures - are LEDs powerful enough to provide more than a romantic-candle-light-dinner light?
(and I live in Brussels as well)
I guess you would have to swap a few more things than IP and port numbers if there is a proper firewall in the way, but I believe that it should be possible - I am not sure if it would work with a standard TCP/IP stack though as you would have to send a SYN/ACK withouth having received a SYN. What would need to happen is the following. You have two computers behind NAT that want to speak to each other, A and B. You have an "enabling" public server C. A and B exchange IP addresses/ports and the ID of the SYN packet you are going to send through C and then agree on a time (i.e. in 0.1s) when you would send SYNs and SYN/ACKs. This should trick the firewall and NAT router into thinking that the SYN/ACKs are the replies to the SYNs sent earlier and open a connection. The problem could be that a firewall could drop outgoing SYN/ACKs without having passed through a SYN before. It might be easier with UDP (which would be good enough for VoIP)? Disclaimer: I don't really know the details here - are there any TCP/IP and firewall wizards around who could see whether this would be possible?
Why would you want to flip?
Here is a link to a high-res mosaic, 3498x3851, TIFF format, 40.4MB:
http://photojournal.jpl.nasa.gov/tiff/PIA04995.tif
And the same picture as a 1.1 MB JPG (still full resolution):
http://photojournal.jpl.nasa.gov/jpeg/PIA04995.jpg
In the SCO cases, I think besides a payoff from law suit and licence sales investors/speculators are betting on a buyout by someone (not IBM though probably). That's why I don't go short (too much risk), because a buyout is possible while a win of the law suit is very improbable.
But will RMS be happy that this artcle is posted to www.linux.com and not www.gnulinux.free?
The relevant organisation is the EEA (European Economic Area) and not the EFTA. EEA = EU + Norway, Iceland and Liechtenstein. EFTA = EEA + Switzerland - EU. Switzerland does not form part of the Single Market, while Norway, Iceland and Liechtenstein do. (I believe they retain the right to veto European Directives, so that they don't apply in their respective countries.)
The information in the article is incorrect. 24 Mbit (or 3 MBytes) of data were received in total over a couple of minutes at a speed of 128 kbps (This was for the first pass of the Odyssey orbiter). 128 kbps relates to the transmission speed from the orbiters to the Deep Space Network on Earth. That's not broadband, but it's not bad either, given the distance! The direct link from the rover to Earth will be much slower with around 12 kbps (but longer time periods of communication can be achieved each day compared to passes of the orbiters). However, this will only happen once the high-gain antenna is operable (which will probably be later today) and the Earth is visible from the rover.