I've added a comment from Sophos's Graham Cluley to the end of the blog post. He/they have been quite responsive, especially given that the free A-V product comes without official support. Apparently I am the only one ever to have reported such a problem with Time Machine.
My backup disk is a Time Capsule, whose internal disk is connected only wirelessly. Probably doesn't count as direct-connected. Does it therefore use sparse bundles?
FYI, I'm not using filevault, just individual files to be backed up... but TM uses sparsebundles in ways I don't begin to understand. One respondent via Twitter suggested that Sophos may have simply been in the process of deleting the entire sparsebundle -- i.e. the entire lot of backups -- when I killed its process. No idea if this is correct. I hope Sophos eventually provides some insight.
Got one. The browsers I tried are Safari 2.0.4 and Firefox 1.5.0.8 -- both of which did pretty well, but not letter-perfect -- and (in Win XP / Parallels) IE 6 and IE 7, both of which were far off the mark.
RTFA. The speculation is that these were not licensing fees. If the media companies collected licencing fees, they would be obliged to share them with the artists. Instead the payments were structured as investments in YouTube, so that the companies could claim capital gains once Google consummated the deal. No artist payments are owed on capital gains. So goes the speculation.
That's the thing: you wouldn't know about mail you failed to receive because Comcast failed to forward it. I run a mailing list that has alum.mit.edu subscribers and I can assure you that for 5 or 6 days, mail sent to alum.mit.edu was not getting forwarded to the Comcast subscribers on my list who are behind those addresses.
>Also, if I understand it correctly, you can really only send an
>encrypted message to one person at a time, because you're
>encrypting it with their public key (so that their private key
>decrypts it). So PGP is not really a solution for, say, mailing
>lists.
Not so. You can encrypt a message to any number of public keys. Any one of them can decrypt it with her private key. If a recipient has the public keys of the others on her keyring, she will be able to see who else can read the message.
I always encrypt outgoing messages to the recipient's key and to one of my own. Sometimes it's nice to be able to review what you've written later.
The cause of the damage, which occurred approximately 100km from Singapore on the ocean floor, could not be confirmed. Possible causes include a ship's anchor or minor earthquake.
Now at last the truth can be told.
http://tbtf.com/pics/subhoe.jpg _______________________________________________ Keith Dawson Layer of ash separates morning and evening milk.
The missing mass needed to close the universe has always been assumed, I've assumed, to exist in the form of either WIMPs or MACHOs (massive compact halo objects) or the Cosmological Constant. Interesting times when evidence for all three is strengthening at once. The current Science News features a solid survey of the unanimity the remarkable idea of an accelerating universal expansion has garnered in just two years -- so much so that the current best-guess value for the CC, the push factor, is engraved on a plaque at the top of the spiral "walk through time" in the new Rose Center (formerly the Hayden Planetarium) in NYC. And convincing evidence for the existence of MACHOs was presented at the recent Atlanta meeting of the AAS. (I'll have links for all these loose ends when the next TBTF issue comes out.)
Slashdot Mirror
I've added a comment from Sophos's Graham Cluley to the end of the blog post. He/they have been quite responsive, especially given that the free A-V product comes without official support. Apparently I am the only one ever to have reported such a problem with Time Machine.
My backup disk is a Time Capsule, whose internal disk is connected only wirelessly. Probably doesn't count as direct-connected. Does it therefore use sparse bundles?
FYI, I'm not using filevault, just individual files to be backed up... but TM uses sparsebundles in ways I don't begin to understand. One respondent via Twitter suggested that Sophos may have simply been in the process of deleting the entire sparsebundle -- i.e. the entire lot of backups -- when I killed its process. No idea if this is correct. I hope Sophos eventually provides some insight.
> Time to get a new computer. [apple.com]
Got one. The browsers I tried are Safari 2.0.4 and Firefox 1.5.0.8 -- both of which did pretty well, but not letter-perfect -- and (in Win XP / Parallels) IE 6 and IE 7, both of which were far off the mark.
Agreed, I changed it.
RTFA. The speculation is that these were not licensing fees. If the media companies collected licencing fees, they would be obliged to share them with the artists. Instead the payments were structured as investments in YouTube, so that the companies could claim capital gains once Google consummated the deal. No artist payments are owed on capital gains. So goes the speculation.
I applied a correction: the article says "vaccine-resistant." It was wrong in the submission and I missed it.
Thanks, I changed this. There really is no perfectly appropriate topic for this story.
Gbit erratum corrected, thanks for spotting.
See Marcel Bresink's utility Temperature Monitor at http://www.bresink.com/osx/TemperatureMonitor.html .
For the first (free) 6 months, you can't change your user ID or password, according to the FAQ in the linked article. They plan to let you, later on.
Here is an easy ay to package up a torrent.
Heh, nice catch, fixed.
Look closely. The block happened last Sunday. Yes, Comcast also blocked the WELL 3 years ago and Declan McCullagh's intervention dissuaded them.
Wonky DNS servers: Tough luck.
See for example OpenDNS, David Ulevitch's startup that aims to give everybody access to superior DNS service for free.
That's the thing: you wouldn't know about mail you failed to receive because Comcast failed to forward it. I run a mailing list that has alum.mit.edu subscribers and I can assure you that for 5 or 6 days, mail sent to alum.mit.edu was not getting forwarded to the Comcast subscribers on my list who are behind those addresses.
Correct, thanks for noticing.
>Also, if I understand it correctly, you can really only send an
>encrypted message to one person at a time, because you're
>encrypting it with their public key (so that their private key
>decrypts it). So PGP is not really a solution for, say, mailing
>lists.
Not so. You can encrypt a message to any number of public keys. Any one of them can decrypt it with her private key. If a recipient has the public keys of the others on her keyring, she will be able to see who else can read the message.
I always encrypt outgoing messages to the recipient's key and to one of my own. Sometimes it's nice to be able to review what you've written later.
The cause of the damage, which occurred approximately 100km from Singapore on the ocean floor, could not be confirmed. Possible causes include a ship's anchor or minor earthquake.
Now at last the truth can be told.
http://tbtf.com/pics/subhoe.jpg
_______________________________________________
Keith Dawson
Layer of ash separates morning and evening milk.
The missing mass needed to close the universe has always been assumed, I've assumed, to exist in the form of either WIMPs or MACHOs (massive compact halo objects) or the Cosmological Constant. Interesting times when evidence for all three is strengthening at once. The current Science News features a solid survey of the unanimity the remarkable idea of an accelerating universal expansion has garnered in just two years -- so much so that the current best-guess value for the CC, the push factor, is engraved on a plaque at the top of the spiral "walk through time" in the new Rose Center (formerly the Hayden Planetarium) in NYC. And convincing evidence for the existence of MACHOs was presented at the recent Atlanta meeting of the AAS. (I'll have links for all these loose ends when the next TBTF issue comes out.)