Slashdot Mirror

The Brains

They do LOVE to gamble!

You DO NOT need to close applications in Android. It's handled automagically by the OS.

It's a hangover habit from the desktop world where you need to close applications when your finished with them. You don't need to even think of it on Android and how it works is rather a refreshing piece of OS design (to the point Apple somewhat copied it for iOS).

There is a lot of misunderstand about how Android multitasks, which is really rather innovative that we could have used in operating systems a long time ago. Read up here: http://android-developers.blogspot.com/2010/04/multitasking-android-way.html

Task killer apps are not necessary and actually work to destablise and lag your phone if you over use them (as I found out the hard way). Learn to let go of the need to control everything and your phone will work faster and crash less often, and you'll have some time and brainpower spared.

when I made a model of an atom in 7th grade science and used spheres for electrons.

But it didn't look as good as her's: http://jeaninallhonesty.blogspot.com/2009/12/how-lizzy-made-atom-model.html.

http://ps3mediaserver.blogspot.com/

Works like a charm, multiplatform (java), transcodes incompatible formats and containers.

The referred-to article talks about the Google App Engine and does not mention the new possibility of creating AI Apps in JavaScript to serve up artificial intelligence to desktop computers, laptops, smartphones and netbooks. The JavaScript AiMind is just a tutorial but fully functional version of a much faster and more powerful AI known as MindForth in Forth for autonomous robots. An AI Mind residing on a JavaScript server flits across the 'Net to take up residence as a client-side AiApp on any stationary or mobile computing platform.

The referred-to article talks about the Google App Engine and does not mention the new possibility of creating AI Apps in JavaScript to serve up artificial intelligence to desktop computers, laptops, smartphones and netbooks. The JavaScript AiMind is just a tutorial but fully functional version of a much faster and more powerful AI known as MindForth in Forth for autonomous robots. An AI Mind residing on a JavaScript server flits across the 'Net to take up residence as a client-side AiApp on any stationary or mobile computing platform.

Yeah there is a nice blogpost linked from the bug with a good explanation. http://muizelaar.blogspot.com/2011/04/webp.html. I was especially interested in :-

"Flickr compresses their images at libjpeg quality of 96 and Facebook at 85: both quite a bit higher than the recommended 75 for “very good quality”. Neither of them optimize the huffman tables, which gives a lossless 4–7% improvement in size. Further, switching to progressive JPEG gives an even larger improvement of 8–20%."

Opera Turbo uses WebP to compress images on low bandwidth connections

http://googlesystem.blogspot.com/2011/04/opera-turbo-uses-webp-to-compress.html

WebP looks better than JPEG at high compression ratios where JPEG has noticeable blocking artifacts.

If you read the article instead of looking at the pictures, the authors point out that this is one situation where the algorithm fails - the method is optimized for edge-finding in hand-drawn 8-bit pixel graphics. In more photo-realistic images like the doom face, especially later games where the graphics were rendered in 3d/hi-res and imported into the game (Donkey Kong Country would be a good example of this), vectorization doesn't work well due to the smoothed anti-aliased graphics.

The main branch of the NYPL uses the same system, albeit more floors that aren't as tall, and human workers handle pick and place.

An original illustration here, sorry for the ugly url: http://2.bp.blogspot.com/_PotguXM3PJk/TKh0YeRyQMI/AAAAAAAAF_c/WiOrMXEWdQc/s1600/nyplstacks.jpeg

Currently, it only supports a subset of the features that JPEG has. It lacks support for any color representation other than 4:2:0 YCrCb. JPEG supports 4:4:4 as well as other color representations like CMYK. WebP also seems to lack support for EXIF data and ICC color profiles, both of which have be come quite important for photography. Further, it has yet to include any features missing from JPEG like alpha channel support.

[...]

Every image format that becomes “part of the Web platform” exacts a cost for all time: all clients have to support that format forever, and there's also a cost for authors having to choose which format is best for them. This cost is no less for WebP than any other format because progressive decoding requires using a separate library instead of reusing the existing WebM decoder. This gives additional security risk but also eliminates much of the benefit of having bitstream compatibility with WebM. It makes me wonder, why not just change the bitstream so that it's more suitable for a still image codec?

WebP, by Jeff Muizelaar.

http://1.bp.blogspot.com/-zSFMapHML0g/TYgH8GrOA3I/AAAAAAAAC8k/roRfUZhfxKE/s320/you_mad_camron.jpg

What planet are you from? 80%? Complete fiction. Vermont Yankee is very reliable, and had, from 2003-2009, an amazing 92.6% capacity factor. Which gives an employee/Mwatt ratio closer to 1.09, which while still slightly higher than the solar plant, isn't particularly bad.

The source for my claim is an open letter from an Entergy executive, being mirrored at the website of Meredith Angwin, who runs the Yes, Vermont Yankee blog.

For more actual *facts* about VY reliability, see this posting at Yes, VY.

In general, nuclear power plants in the U.S. have had an *industry average* of over 90%. That's not a cherry picked record for an individual plant - that's the *average* capacity factor. There are certainly some things to be worried about Nuclear plants, in terms of risks and costs, but reliability just isn't one of them. Let's stick to real problems, instead of making up fake ones.

As for number of employees per MW at nuclear plants, there is probably room for improvement there, with newer designs. However, I don't see that 650 employees for 620MW seems like a particularly *bad* ratio. As mentioned above, it's less than 1.09 empl./MW, so it's in the same general ballpark as the solar plant.

Pic here: http://1.bp.blogspot.com/_mpBGa4P5jUo/TL42QnvUzkI/AAAAAAAAFuk/B4g6Bp2czH8/s1600/altrusia7.jpg

Greetings Professor Falken

A strange game.

The only winning move is
not to play.

Greetings Professor Falken

A strange game.

The only winning move is
not to play.

South and east, eh? And minimal hiking?

  * The pictures I've seen of the beaches near Vík are stunning. :)
  * If you're going to rent a 4x4, you can drive through the Landmannalaugar area.
  * Skógafoss is amazing, and you can camp right by it. Of course, you said you're already going up the Skóga....
  * Svartifoss is a bit more out of the way and smaller, but beautifully framed by columnar basalt cliffs.
  * I've heard the Morsá valley recommended, and it definitely looks like an easy hike, although the pictures I've seen don't impress me as much as elsewhere.
  * Jökulsárlón is an obvious candidate, and kind of hard to avoid anyway ;)
  * I assume you're not planning to get on top of Vatnajökull ;)

But you probably already know that stuff ;) Ooh, actually, here's something you may or may not know: in August, this happens at Jökulsárlón. :) Wish I could be there for it, but I had to choose between that and Bræðslan.

Are you serious? Do you really think that the Windows platform can't scale to handle a few million users and a few million hits/day? Depending on how CPU and database intensive his app is, it's likely that he could scale that far on a single physical server. 8 cores + 8GB of RAM will take you far, even on Microsoft.

Appengine lets someone who'se never heard of load-balancing or database replication do the Royal Wedding website:
        http://googleappengine.blogspot.com/2011/05/royal-wedding-bells-in-cloud.html
If you're telling me you'd be comfortable hosting that on a Windows VM... well, let's just say you wouldn't have enjoyed the day much!

We have no idea what this guy is building, but needing that sort of scalability is not unheard of. I myself have a neat application that could conceivably be mentioned on Oprah, and if it happens I'll be popping champagne corks instead of bloodvessels - I'm on Appengine.

"can replace in use files" - I'll try to be reasonable here and ask this out of curiosity: I haven't fully used linux in years - are you telling me you can update the kernel live while logged in and without rebooting? If that's the case that's just damned impressive.

Yes, you can.

Is that the basis for this imagery? There is however a left behind story that describes the the VelociRapture very differently.

The 1987 Montreal Protocol banning chlorofluorocarbons (CFCs) is considered a textbook case where science and responsible governance lead to a landmark treaty for the benefit of the Earth and all its inhabitants. How often does that happen?

At about the time that the DuPont patent on Freon(TM), the most widely used CFC refrigerant in the world, was expiring the mainstream media picked up on otherwise arcane scientific observations and hypotheses about ozone concentration in the upper atmosphere near the poles.

There resulted an international mobilization to criminalize CFCs and DuPont developed and patented a replacement refrigerant that was promptly certified for use.

A Nobel Prize in chemistry was awarded in 1995 for a laboratory demonstration that CFCs could deplete ozone in simulated atmospheric conditions. In 2007 it was shown that the latter work may have been seriously flawed by overestimating the depletion rate by an order of magnitude, thereby invalidating the proposed mechanism for CFC-driven ozone depletion [3]. Not to mention that any laboratory experiment is somewhat different from the actual upper atmosphere. Is the Nobel tainted by media and special interest lobbying?

[3] Nature 449, 382-383 (2007).

And putting on the paranoid hat, this may be a driving force behind it:

HBGary sold devices to the government so that they could perform the same sort of trick. -- Errata Security

When you have inflated money due to the large amount of worthless paper, it is still hard to get anyone motivated to make the bread. In hyperinflation the wheelbarrows of printed money came from somewhere. As it was mass produced, and was less accepted, unemployment ran high, money was worthless, and nobody made the bread to put on the store shelves.

Some bread was made but it was barter as another economy trading goods and services grew in favor instead of truckloads of money that could not buy anything.
http://junksilver.com/history-of-the-collapse-of-fiat-paper-currencies.html
http://3.bp.blogspot.com/_fKeKEf4ylyU/Sal9gmRdA2I/AAAAAAAAACU/O3JO8HXtLZE/s1600/Weimar-Republic-Lady-Using-Money-To-Heat-Home-1923.jpg

May is Zombie Awareness Month.
( Well, actually, there are many unofficial dates set, so I observer all of them! )

Zombie Awareness Day is a great time to check your friends and family for bite marks and other tell tale "infected" behavior, and to review & revise your Zombie Plan (Think Fire Escape Plan -- Except that you're prepared to keep running for months after you safely exit the premises).

Which brings me to my next point: It's time to make sure you have your Zombie Preparedness Kit in order -- It's basically a Hurricane or Earthquake preparedness kit, with more shotguns and shells.

Remember -- "Shoot it in the head, it stays dead.", and have a Happy Zombie Awareness Month!

there will be a fork. besides, it isn't like everyone is sticking with the gnome interface.

http://tuto4log.blogspot.com/2011/04/ubuntu-411-arrives-with-unity-new.html

Or it could be that cultural differences matter much more than whatever is being done in the school.

Iowans score on par with asians on those test results. So do children of Scandinavian decent living in the US in general.

Immigrant families in sweden score much lower.

But it couldn't be that only the rich and successful parts of China take part in that test, or that Finland (or for that matter, most of the US) has a culture that is much more functional than the lower-income inner-city culture dragging our test scores down.

We make a big mistake when we act as though education policy is the only, or even the largest, controlling variable here.

http://super-economy.blogspot.com/2010/12/amazing-truth-about-pisa-scores-usa.html

Fortunately the accused in our system don't have to prove anything, just convince the jury that there's reasonable doubt. If it boils down to a he-said/she-said situation, that shouldn't be too hard. If there's further evidence of sexual entrapment, even easier.

But, no matter, he'll have been replaced at the IMF even before a pre-trial hearing, so the goal will have been met and it doesn't matter what the judicial outcome is.

Shame that the presumption of innocence does not apply to rape in practice.

Alternate DOS extenders.
http://maximumhoyt.blogspot.com/2008/12/dos4gwexe-version-201a-and-alternative.html

The most useful appears to be DOS/32A, a drop-in replacement for DOS4GW.EXE .
http://dos32a.narechk.net/index_en.html

Looks like my memory got a few details wrong:

http://winstonscifi.blogspot.com/2010/05/synopsis-for-secret-of-martian-moons-by.html

steveha

something as vaporous as 'there might be an earthquake in the next few days' isn't going to change routines

Please tell me you're kidding.

http://www.abovetopsecret.com/forum/thread701933/pg1

http://www.cnkeyword.info/taiwan-14-earthquake-predicted-wave-triggered-evacuation-helicopter-apron-shift/

http://gardencityquake.blogspot.com/2011/03/false-earthquake-prediction-sparks.html

People are willing to evacuate cities based upon the predictions of this guy: "He is co-author of Pawmistry: How to Read Your Cat's Paws, which teaches ways to read a cat's mood and "explores the psychic influences that numerology and the zodiac have on your cat"."

Hopefully a scientifically backed prediction would have at least as much effect.

"N.S.A. has a five-thousand-acre campus at Fort Meade protected by iris scanners and facial-recognition devices."--this is false as of a couple years ago~2009. They were experimenting with facial recognition (had some contractors working on it) but it was by no means widely implemented. There was only one office in the research building (R) of the ft meade campus that had an Iris scanner as far as I know and that was also just experimenting. "The electric bill there is said to surpass seventy million dollars a year." Partially because the NSA mismanaged the power supply. Building compute farms without figuring out where to get the power to run them (while I was there they would have regular power outages). When google builds a data center the first thing they do is figure out where they will get the power (usually building near the supply ie a hydroelectric damn). Another thing to note is that in late 2007 NSA started looking at Hadoop which is an open source implementation of some of googles infrastructure, map-reduce in particular. Who knows how much google cooperates with the NSA, at least a little. I was at a meeting at the NSA that included Vint Cerf, a prominent google employee, and they were using a google search product internally. In general it is pretty hard to not witness fraud at the NSA. Here are some examples: http://natsecurityeb.blogspot.com/2010/10/top-secret-america.html

Quoting Linus Torvalds:

"So every time I see some piece of medical research saying that caffeine is good for you, I high-five myself. Because I'm going to live forever."

http://torvalds-family.blogspot.com/2010/08/13744-supplied.html

Not only is it far from cheap, the chemicals used in the resin are far from cheap. Previous DIY versions (e.g. http://3dhomemade.blogspot.com) are much cheaper to build, though to my knowledge there has yet to be a breakthrough in finding a cheaper (near)UV-activated resin with the right characteristics.

This is remarkably similar to another 3D printing project that I've been watching closely:

http://3dhomemade.blogspot.com/

This guy is using a DLP projector, some custom software, and a working surface that raises very slowly out of the printing resin.

From Youtube:

http://www.youtube.com/watch?v=snOErpOP5Xk&feature=player_embedded

There has been quite a lot of speculation about the project, with most of it centered on the resin that he's using (which he hasn't divulged yet). I've done some researching and found that UV activated masking materials may be a likely candidate as they cure quickly and form thin layers.

If it can be developed for under $1,000 (excluding the projector), I think it would be very successful.

you can't do search-and-replaces, pipe the registry through other utilities to make more advanced changes

I may have misunderstood what you mean, but you can do all of that, using Powershell. It's quite easy too. Powershell contains what's called a "registry provider" that allows you to enumerate though the registry items and pipe them through other utilities; see here how to use the registry provider. Look here, here or here for examples of piping registry items through other utilities

http://3dhomemade.blogspot.com/

Which is a homebrew version of EnvisionTEC's commercially available machines. Which makes this essentially the same thing as the linked machine but another person doing it.

I wrote about my preferred dual-monitor setup a while back as a guest editorial at: http://onyourdesktop.blogspot.com/2007/07/david-mertz.html

That's still pretty much what I like. I wish the screenshot there was from the work machine I describe in the article, with dual 30" screens. Sadly, I haven't had such nice desktop real estate elsewhere (neither at home nor other workplaces). But what I do typically work with in my own office space nowadays (I'm a consultant, so it's really my own space) is still two screens, though not identical in dimension. I use a (company provided) laptop hooked up to an external 24" monitor (1920x1200--those widescreen ones with only 1080 vertical pixels feel like they rob me of important vertical space). On this main consulting setup I keep a pretty fixed set of apps open... sadly, the work laptop is Windows 7, and I can't choose otherwise. But my setup hides most negatives of that OS.

To the left, on the laptop screen, everything maximized:
  * Email client (usually in front)
  * Version control GUI

To the right, on the external screen:
  * Code editor (jEdit) maximized. This allows two full panes of code, and one
        slightly smaller one for file navigation, project manager, search results, etc
        (my editor tabs between different functions in the utility pane). Each code
        pane is about 85x65 in a reasonably large font. My editor also lets me hide
        all the frame elements, title bar, etc, which removes the look of Windows and
        gives me a couple extra rows of code.
  * Web browser, full vertical, but only about half the total screen with. Lots of tabs
  * Two side-by-side SSH sessions, each one about 90x70 characters. These connect
        to the real machines where I do work. Often I run vim in these session to edit
        code on the remote machines, but also to run test commands, launch compilation,
        etc.
  * Sometimes a chat window or two that use the full monitor height.
  * Sometimes a PDF viewer or two, usually maximized and two-page display

Obviously, I have to switch focus sometimes on the right (external) monitor. But most times I am just looking at the two large SSH windows. It's a bit disjunctive what I edit in my local text editor versus what I edit in my two "panes" of vim on the SSH terminals; but either way I can see a similar two full screens of code to compare visually, which is really useful (e.g. one I use to look at the code of the supporting library while in the other pane I write the code that calls into it; or I am working on two related scripts and seeing both next to each other helps synchronize changes).

Some guy whose blog appears to have been deleted alleges in a newsgroup posting (archived in Google Groups) that the U.S. Government is behind the Sony data breaches. Not sure I believe that, but it could explain Japan's stance...

Actually, I agree with this alot.

And the alot will be happy to hear that.

You know that many chiropractors deny the "germ theory" of disease, right? Eg: http://chiropracticwellnesscenter.blogspot.com/2010/08/your-health-and-germ-theory.html

He holds up as examples the BBC and PBS - they just seem "fair and impartial" to him (and millions of other, left-leaning folks).

The CBO is biased to - but not by their own hands, they are limited to "cost" bills and make projections fed them by partisan hacks - they can't independently go out and gather their own data they have to base their projections on the mis-information they are fed by the politicians and their staff.

If President Clinton thinks the Government should play an active role in the truth, maybe it could start by adopting a rule to only tell the truth? That would be a much better start, IMHO.

This article by some nuclear engineers at NC State is an excellent, fact-based breakdown of what the effects are of the Fukushima accident, with known numbers to date.

Bottom line: Three cancers.

Three cases of cancer that would not otherwise have occurred, and this is using the (very conservative) linear-no-threshold assumption.

Others in this thread have been bleating about how bad nuclear power accidents have been. The following quote from the UN's final report on the Chernobyl accident (a summary can be found here ) doesn't support their claims:

"Apart from the increase in thyroid cancer after childhood exposure, no increases in overall cancer incidence or mortality have been observed that could be attributed to ionizing radiation. The risk of leukemia, one of the main concerns (leukemia is the first cancer to appear after radiation exposure, because of its short latency time of 2 to 10 years), does not appear to be elevated, even among the recovery operation workers. Neither is there any proof of other non-malignant disorders that are related to ionizing radiation. However, there were widespread psychological reactions to the accident, which were due to fear of the radiation, not to actual radiation doses."

People's fear is very real and important. But it's not substantiated by facts.

Old RAM advice is old.

Just do a "free -mto" or open up the perf monitor in taskman.exe . With all your common applications open, if you still have more than a few megs of free memory (instead of cached), then you probably have too much RAM.

These days, I would modify that to say RAM before SSD. You can typically load up on another 8GB+ of RAM for less the the cost of the cheapest SSD, and it will have a more profound effect on the apps you always have open. RAM is still more than 100x faster than even the high-end SSDs, but SSDs aren't necessarily more than 10x faster than a decent cheap hard disk, even with lots of small reads provided you use readahead to preload a lot of your HD data to RAM, and of course migrate /tmp to tmpfs or something.

http://trumblings.blogspot.com/2010/11/using-readahead-to-speed-up-disk.html
(and for recent Fedora, Ubuntu/Debian, etc. your OS is already using readahead to boot fast)

If you never shutdown your laptop or desktop and just put them in and out of suspend, this cache is always maintained in RAM where most of your critical OS and applications never expire from, so you're kinda not benefiting from your SSD as much as you expect anyway. Maybe if you used your SSD for swap, but people don't tend to like to do that ;-)

Well, informing the public is a start. Buy Kindle, buy books for Amazon, and be forever dependent on Amazon's gadgets to access the books you purchased for almost the same price as the dead tree versions! I'm a big fan of eBook readers - I hate reading on a backlit LCD. When I was shopping for an eBook reader, I carefully considered all options, and ended up with... WAIT FOR IT... SONY - yeah, that's right. No DRM on their online bookstore, and it reads everything you throw at it - PDF, .epub, text, even MS Word files. Fully supported by the FLOSS Calibre book management software... Comes with SD card slot, replaceable battery, touch screen and the FULL Oxford Dictionary (both AmE and BrE). After using it for some time, I wrote a short review, and explained in details why Kindle is a bad idea. I also happened to recommend piracy, for an entirely different reason (not that my review is intended to Vietnamese students and complete noobs).

staples easy tech is just high pressure sales with real techs forced out.

The thing is at stapes you better sell.

http://paulrepair.blogspot.com/2009/04/stay-away-from-staples-for-computer.html

http://www.complaintsboard.com/complaints/staples-c276403.html

http://consumerist.com/2008/03/staples-tries-to-charge-senior-citizen-390-for-basic-computer-repair.html

http://consumerist.com/2008/09/why-i-quit-staples-easy-tech.html

http://consumerist.com/2009/12/i-always-look-forward-to.html

Canada is taxing the memory of those who don't have Alzheimers. If you don't pay up, expect a visit...

only a limited number the most popular devices get tested and approved because after that the cost/benefit ration drops

This argument doesn't hold water, because two out of five supported devices - namely, Nexus One and Nexus S - are very far from being the most popular.

For example, here are downloads of Google's own app Androidify from the Market as of March 13 this year. Note that there is no "Nexus" in the top 10 anywhere, yet the #3 phone is not supported.

Netflix has chosen the first option which is the most sensible right now on Android if you still want to do some quality assurance.

Numerous other applications, made by companies far smaller, manage to handle this right while retaining excellent app quality. If Netflix is having problems here, they should hire better devs or something.

Relax :) (blog post from one of the head developers of KDE)

KDE5 will not break the world like KDE4 did. Just as Qt5 will not break the world.

http://aseigo.blogspot.com/2011/05/relax.html
http://aseigo.blogspot.com/2011/05/qt5-kde5.html

KDE5 will not break the world like KDE4 did. Just as Qt5 will not break the world.

http://aseigo.blogspot.com/2011/05/relax.html
http://aseigo.blogspot.com/2011/05/qt5-kde5.html