Forgot your password?

Comment: Re:This is why digital sucks (Score 1) 220

by lgw (#47911199) Attached to: Ask Slashdot: What To Do After Digitizing VHS Tapes?

Stone tablets from Babylon are still readable because analog degrades gracefully.

Want real long-term storage? Write them to analog tape again.

Metal punch tape - chosen by the military to survive nuclear apocalypse and still be human readable without tools. Much denser than cuneiform, too.

Comment: Re:Final Cut Pro library (Score 4, Insightful) 220

by lgw (#47911131) Attached to: Ask Slashdot: What To Do After Digitizing VHS Tapes?

For pro video editing - which is to say lots of content that frequently changes - tape backup still makes sense. There's still no better way to archive large amounts of data, although 2.5 TB (IIRC) tape size for the latest LTO has fallen behind, and the next gen isn't due for likely a year.

But the one-time cost for tape drives is pretty steep. If you're going to use many tapes each month, it's worth it. Heck, I'd say even at 10 TB of new data a month being archived, there's no better way. But for, say, 10 TB of fixed data that just needs to be archived once, it's overkill.

Buy a few HDDs, keep their shipping containers, make a backup and ship them to a friend in a different state. Repeat yearly. That's the economical way. Eventually it will all fit on a single drive, after all (aren't there leading edge 10 TB drives already?), and so you're looking at ~$10/month long term (heck, no matter how much fixed data, eventually it will fit on one drive and cost about that much).


Comment: Re:why? (Score 1) 177

by TheRaven64 (#47907301) Attached to: Oculus Rift CEO Says Classrooms of the Future Will Be In VR Goggles
Add to that, about 10-20% of the population get motion sick using the kind of VR in Oculus Rift (myself included - I can use it for 2-5 minutes, depending on the mode). It's ludicrous to imagine building a school that would exclude 20% of the potential pupils on some random criterion. You might as well make schools that didn't let in gingers...

Comment: Re:intel atom systems keep 32 bit systems around (Score 1) 127

by TheRaven64 (#47907067) Attached to: Chrome For Mac Drops 32-bit Build
Apple already ships 64-bit ARM chips and a lot of other vendors are racing to do so. The Android manufacturers that I've spoken want 64-bit for the same reason that they want 8-core: It's a marketing checkbox and they don't want to be shipping a 32-bit handset when their competitor is marketing 64-bit as a must-have feature. ART is in the top 10 worst-written pieces of code I've had to deal with and is full of casts from pointers to int32_t (not even a typedef, let alone intptr_t), but it should get a 64-bit port soon.

Comment: Re:The ones I witnessed... (Score 1) 127

by TheRaven64 (#47907057) Attached to: Chrome For Mac Drops 32-bit Build
64-bit is here for a while. A lot of modern '64-bit' CPUs only support 40-bit physical addresses, so are limited to 'only' 128GB of RAM. Most support 48-bit virtual addresses (the top bit is sign extended, so all 1 or all 0 depending on whether you've got a kernel or userspace address), limiting you to 'only' 32TB of virtual addresses. If RAM sizes continue to double once every year, then it takes another year to use each bit. We currently have some machines with 256GB of RAM, so are using 41 bits. 64 bits will last another 23 years. RAM increases have slowed a bit recently though. 10 years ago, you always wanted as much RAM as possible because you were probably swapping whatever you were doing. Now, most computers are happy with 2GB for programs and the rest for buffer cache. As SSDs get faster, there's less need for caching, but there might be more need for address space as people want to be able to memory map all the files that they access...

Comment: Re:It's not Google's fault. It's Mozilla's. (Score 1) 127

by TheRaven64 (#47906847) Attached to: Chrome For Mac Drops 32-bit Build

The real problem for Firefox is not the interface changes that people like you whine about, it's mobile. Now 30% of traffic is mobile and Firefox doesn't have an app for any Apple mobile devices and is effectively excluded from Android by Google's Microsoft-like illegal anti-competitive licensing deals with manufacturers (you can get the app, but it's not preloaded and only a few geeks ever would).

Huh? It's in the Google Play Market and is no harder to install than any other app. Once it's installed, the first time you click on a link from another app you're asked to choose the app that will handle links. I fall into the geek category (and so installed it from F-Droid, not Google Play), but found it trivial to switch to Firefox on the mobile. I mostly did because Chrome has spectacularly bad cookie management and I'd been trying to find a browser that did it better. Early Firefox ports were as bad, but now it's quite nice and with the Self Destructing Cookies add-on does exactly what I want.

The mobile is actually the only place I use Firefox...

Comment: Re:Problem? (Score 2) 267

I just come back to "probable cause". Any search without a warrant is bad, and to get that warrant you should need to show that more than half of those you search have the specified contraband. That's what "probable" means, after all.

Bayesian reasoning tells us that's a remarkably high bar to clear based on any sort of profile, but it's technically possible. If, say, you have good evidence that more than half of those who visited Silk Road have illegal drugs in their house right now, then, OK, that's a legit reason to search the houses of everyone who did.

But most profiling and broad searches are closer to 0.05% than to 50%. Search all the computers in the state and find one guilty? What percentage is that? Stop 1000 people at a sobriety checkpoint for every drunk you find? Well, that's a bit less than half, now, isn't it. Search people who fit a profile because they have a one in a million, instead of 1 in 100 million chance of being a terrorist? "They're 100 times as likely if they fit!" Yeah, well, 1 in a million is less than 1 in 2, so keep working on that profile buddy.

Probable cause. It's a simple concept.

Comment: Re:Where is the misuse of military equipment charg (Score 1) 267

Well, if civilian rules of evidence were in play, the evidence should still be thrown out - an overbroad search is an overbroad search. Even though might have found the same evidence with a narrow search, you didn't. But then, I have no clue what the rules of evidence are for the UCMJ, and it's a different world than civilian law. (And of course in the civilian world, they'd just use parallel construction to falsify the origin of the evidence.)

Comment: Re:When the cat's absent, the mice rejoice (Score 1) 267

Everyone but you is construing your post to mean that the government investigators was OK to exceed his authority because child molesters are scum. When you call enough people idiots for misunderstanding you, you should start to think that you were perhaps unclear.

Or as the old saying goes "if everyone you see is an asshole, look in the mirror".

Comment: Re: No, no. Let's not go there. Please. (Score 1) 852

by TheRaven64 (#47902613) Attached to: Why Atheists Need Captain Kirk
I recall reading some years ago that there are two kinds of atheists:
  • Those that disbelieve all religions.
  • Those that disbelieve all except one religion.

For some reason, people in the second category describe themselves as 'religious'. And yet you'll be hard-pressed to find, for example, a Christian who requires the same standards of evidence for the non-existence of the Norse, Egyptian, Greek or Hindu gods as he requires that an atheist from the first category provides for the non-existence of the Abrahamic god.


US Patent Office Seeking Consultant That Can Stamp-out Fraud By Patent Examiners 123

Posted by Soulskill
from the stamping-out-the-rot dept.
McGruber writes: A month after Slashdot discussed "Every Day Is Goof-Off-At-Work Day At the US Patent and Trademark Office," the USPTO issued a statement that it is "committed to taking any measures necessary" to stop employees who review patents from lying about their hours and getting overtime pay and bonuses for work they didn't do.

USPTO officials also told congressional investigators that they are seeking an outside consulting firm to advise them on how managers can improve their monitoring of more than 8,000 patent examiners. The Patent Examiners union responded to the original Washington Post report with a statement that includes this line: "If 'thousands' of USPTO employees were not doing their work, it would be impossible for this agency to be producing the best performance in recent memory and, perhaps, in its entire 224 year history."

In related news, USPTO Commissioner Deborah Cohn has announced plans to resign just months after a watchdog agency revealed that she had pressured staffers to hire the live-in boyfriend of an immediate family member over other, better-qualified applicants. When he finished 75th out of 76 applicants in the final round of screening, Cohn "intervened and created an additional position specifically for the applicant," wrote Inspector General Todd Zinser in a statement on the matter.

Comment: Re:Really? (Score 1) 287

by Roger W Moore (#47888505) Attached to: German Court: Google Must Stop Ignoring Customer E-mails
I expect that part of what colours my experience is that the companies I typically have to call most often for customer support are the ones with the longest waits and slowest response times (that's if you even get one). I suspect that there may be a correlation. The couple of times I've had to contact Apple over the past several years they have been amazingly fast but try talking to Shaw Cable or Air Canada (other than to book a flight) and make sure you can sit by the phone for an hour or more and you can forget email because they certainly seem to!

Comment: Re: ...and say what exactly? (Score 1) 362

by Roger W Moore (#47888423) Attached to: BBC: ISPs Should Assume VPN Users Are Pirates
That's what the summary says but the summary is WRONG. Read what they actually said in the article. Deliberately mis-stating what has been stated to make it ridiculously inflammatory is counterproductive and makes any criticism easy to dismiss. All the BBC says is that pirates use VPN. The troubling part of what they say is not this but that ISPs should act on any wild accusation they get to cut, or restrict, people's internet access. There is plenty to criticize there without the need to make up crazy stuff that was never stated anywhere.

When you are working hard, get up and retch every so often.