Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?
DEAL: For $25 - Add A Second Phone Number To Your Smartphone for life! Use promo code SLASHDOT25. Also, Slashdot's Facebook page has a chat bot now. Message it for stories and more. Check out the new SourceForge HTML5 Internet speed test! ×

Comment Re:Proofreading... (Score 1) 135

It appears the comment stream got merged with the article that discussed Mozilla's use of SeaMicro servers. Oddly enough, the Mozilla article now has no comments...

This ^ . I was replying to the low-powered servers article, not anything about some technologist named Kevin. I was referring to the fact that the article synopsis misspelled the word "at", instead saying "st", hence the intentional typo in my post.


Submission + - Who Wants to Live Forever?

Hugh Pickens writes writes: "Reuters reports that if biomedical gerontologist Aubrey de Grey's predictions are right, the first person who will live to see their 150th birthday has already been born and the first person who will live to 1,000 is likely to be born less than 20 years after the first person to reach 150. De Grey, chief scientist of the non-profit California-based "Strategies for Engineered Negligible Senescence" Foundation sees a time when people will go to their doctors for regular "maintenance," which by then will include gene therapies, stem cell therapies, immune stimulation and a range of other advanced medical techniques to keep them in good shape. "I'd say we have a 50/50 chance of bringing aging under what I'd call a decisive level of medical control within the next 25 years or so," says de Grey. De Grey's ideas may seem far-fetched, but in 2005 MIT Technology Review journal offered $20,000 to any molecular biologist who showed that de Grey's SENS theory was "so wrong that it was unworthy of learned debate" — a prize that has never been won. But for some, a world composed of healthy, powerful, composed, careful people in their seventies, eighties, nineties, or hundreds, or the prospect of living for hundreds of years is not particularly attractive, as it conjures up an image of generations of sick, weak old people and societies increasingly less able to cope. "This is absolutely not a matter of keeping people alive in a bad state of health," says de Grey. "This is about preventing people from getting sick as a result of old age. The particular therapies that we are working on will only deliver long life as a side effect of delivering better health.""

Submission + - Secret Service Raids Apple Store Webcam 'Artist' (

Endoflow2010 writes: The US Secret Service has raided the home of an artist who collected images from webcams in a New York Apple store.

Kyle McDonald is said to have installed software that photographed people looking at laptops then uploaded the pictures to a website.
Writing on Twitter, Mr McDonald said: "@secretservice just stopped by to investigate [web address removed] and took my laptop. Please assume they're reading any e-mails you send me."

Kyle McDonald's images were uploaded to a page on the blogging site Tumblr. In the description of People Staring at Computers, the project is described as: "A photographic intervention. Custom app installed around NYC, taking a picture every minute and uploading it if a face is found in the image.

"Exhibited on site with a remotely triggered app that displayed the photos full screen on every available computer." No arrests had been made in the case as of 8 July.


Submission + - Nominations for the 2011 SysAdmin Awards (

davidu writes: "We SysAdmins never get the love. We often save the day and because we do, nobody notices or even knows what we did.

That's why July is SysAdmin Appreciation Month and nominations for the 2011 SysAdmin Awards end on July 12th. Nominate yourself, or your favorite SysAdmin hero to receive recognition and an award."


Submission + - How to stop your cellphone voicemail being hacked (

An anonymous reader writes: Illicit access to voicemail messages is making headlines in the UK, with the popular "News of the World" tabloid closing down as a result of its involvement in the scandal.

Fortunately, the ways in which mobile phone voicemail can be illegally accessed are well known, and even regular users can take steps to prevent the risk of being compromised.

Mobile security expert David Rogers proposes that cellphone service providers should consider shutting remote access to voicemail down entirely, as a security measure.


Submission + - Snow Falls on The Most Arid Desert on Earth (

crackspackle writes: The Atacama desert region, a vast expanse of land stretching 600 miles along the Pacific coast of South America from Peru to Chile, is know as the dryest region on earth, receiving only .04 inches (1mm) of rain per year. Many weather stations located in the region have no recorded precipitation during their existence. Sterile from the lack of rainfall, sparsely inhabited, and virtually free from electromagnetic and light interference, the desert hosts several major astronomical observatories. The other-worldy location is also popular among sci-fi film makers, and is a prominent test site for NASA's planned Mars mission. This week, the Atacama received 32 inches of snow, stranding motorists along the Pan-American highway and other roads, prompting numerous rescues. Footage of the snow is available on the BBC

Comment Re:Fictional or real (Score 1) 722

This is what my company's official internal wiki "Server Naming Guide" (all server name examples are real, or once real, servers; wiki links removed for posting to Slashdot) says on the topic: (I think it's pretty funny...)

System Name
  • Overall names for systems should be chosen based on two criteria: choosing a bite-sized name (when visualized in terms of bites of alphabet soup) or, if the bite is too large, selecting an acronym that may or may not mean anything or communicate the purpose of the server.
    • A high-quality example of the acronym selection process can be found in otgsrvr; as the systems installed on this server have not worn the “OTG” name since purchase by EMC in 2001 (for those counting, the server was purchased in May 2007). For interpretation of this name, provides this handy reference:
      • OTG-On-The-Go (USB 2.0 specification)
      • OTG-Over the Glasses (goggle type)
      • OTG-On the Ground (sales)
      • OTG-Off the Ground (gaming)
      • OTG-Operations and Technology Group (Microsoft)
      • OTG-Oficina de Treball de la Generalitat (Catalonia)
      • OTG-Off the Grid
      • OTG-On The Green
      • OTG-Old Timers Guild
      • OTG-On The Grid
      • OTG-Originating Trunk Group
      • OTG-Officer Training Group
      • OTG-operational target graphic (US DoD)
      • OTG-Operational Tactics Guide
      • OTG-Operations Tactics Group
      • OTG-OTH (Over-the-Horizon) Targeting Gold (message format)
      • OTG-Older Than God
      • OTG-Oven Toast(er) Grill
      • OTG-Optimal Treatment Guidelines
      • OTG-Original Tool Gangster
    • Going forward, otgsrvr is to be considered in terms of toaster ovens.
  • One good option for acronym selection is to select some letters that seem like a good server name, then think of something to ostensibly link the device’s purpose to those letters (ie bum)


  • If there is no test server, a system is in production unless marked otherwise (ie cognos)
  • If there is a test server, a production system must be marked as productional (ie health-prod-app), except when it isn’t (ie crystalsrvr)
  • If there is no test server, but you want to make extra sure that a server is in production, it can be marked as productional (ie gis-prod-web)
  • Generally, the ‘prod’ label should be used liberally; you never know when someone will see a device named something like ‘vcsrvr’ and say, “hey it doesn’t say prod, so it should be a good place to put databases for unrelated systems!”
    • You can bet that wouldn’t happen to ‘prod-vcprodsrvr-prod-reallyprod’


  • Test systems should either be marked as test, not marked as test, or marked as test in a cryptic and nonstandard way. This can be determined through the roll of a standard 6-sided die;
    • 1-3: Add ‘test’ to the server name, replacing ‘srvr’ where present. Whether to add a hyphen or not before the word is discussed in Appendix B. (ie health-test-db, otgtest)
    • 4: Don’t mark the server whatsoever. (ie oraclesrvr)
    • 5-6: Select a marking style at random! (ie t-websrvr01, test-db05)


  • In all cases, a dev server should be marked as either production or test. The fact that it is a development server should remain a closely guarded secret.

System Function Suffixes

  • A system’s purpose should be appended to the end of the system’s name, but only in the following circumstances:
    • There are multiple systems that would otherwise have the same name, and we don’t feel like using numbers (ie icris-img)
    • Names that would otherwise fall within the netbios 15 character limit, and we would rather push them over the limit and tempt fate (ie psoft9-prod-web2)
    • When we can’t think of anything better (ie so-db)
  • The suffix used should be decided based on one of two factors:
    • Whether the name is descriptive of the purpose (ie real-test-db)
    • -OR-
    • Whether the vendor has some nonsense name that deceives about the server’s true behavior (ie fidlar-sec)

Appendices Appendix A: Numbering

  • Servers can be numbered according to several different methods:
    • Sequential numbering based purely on server age (ie intranet2)
    • Sequential numbering based on location (ievm-weld7)
    • Sequential numbering that goes back to fill in gaps left due to decommissioning of servers (ie printsrvr01)
    • Numbering based on the version of the software installed on the server (ie banner8-prod-db)
    • Sequential numbering based on the version number for the first server, then increasing that number for additional servers despite not increasing the software version number (ie websense8)
    • Adding arbitrary numbers to the end of the server name regardless of need (ie test-db05)
  • Digit count selection
    • The number of digits used should be either too low for the number of servers that may end up in a system (ie vm-weld9) or far too high to ever possibly need (ie dmzroot01)
    • When there is an established name that is having devices added to it, feel free to change the number arbitrarily (ie websrvr01->websrvr3) or substitute a lowercase “O” for one of the digits. (ie dmzrooto1)
  • The number 1 is implied in a system name (ie ciscoacs4srvr) or is not implied (ie imagesrvr01) depending on the flip of a coin.
    • Heads; 1 is implied
    • Tails; 1 needs to be specifically stated

Appendix B: Hyphenation

  • Hyphens should be used to separate parts of server names, especially when doing so will push a name over the 15 character netbios limit (ie psoft9-prod-web2)
  • Hyphens should not be used where it would appear to make the names appear systematic and standardized – the goal is an ordered mess (ie accelatest-app)

Appendix C: “srvr”

  • When a server name is too small of a bite (yes, the alphabet soup metaphor again), ‘srvr’ can be used to beef up the system name. This is not a ‘suffix’ in the terms described above (ie bannersrvr-db)
  • In all cases, it will be replaced by the word ‘test’ if that (the full word, mind you; ie t-websrvr03) is added to the server name, but by itself does NOT imply that the server is production!! (ie oraclesrvr)
  • When the server name ends with “s”, the first letter of ‘srvr’ may be contracted into that “s” (ie wsusrvr), or not (ie iissrvr)

Comment Re:Is happyed a word? (Score 2) 55

Looks like someone did a global-search-and-replace on "happen" with "happy."

Reference Jonathan's answer to question five:

I'm doing this because I'm tired of my imagination being beyond what I'm able to make happy, and I'm in the fortunate position of being able to pay for it myself.

And question ten:

That part of it had nothing to do with me, so I knew there was nothing I could do to make it happy again.

All of those should be "happen" IMHO.

Comment Clark's 3rd Law (Score 1) 1486

"Any sufficiently advanced technology is indistinguishable from magic."

Give it time. Eventually, we'll have evidence in front of us that quantum mechanics, dark matter, dark energy, etc. do indeed work as predicted by science-- because we'll have the products that rely on those phenomena in front of us. Does it matter that we can't prove that they are working such-and-such way? No, not really. If the dark matter propulsion system on your space ship is propelling you across space, either you admit that you have evidence of dark matter, or you believe in magic.

Comment Re:How much is your commute/time worth? (Score 1) 615


I was telecommuting out-of-state with no benefits for almost 4 years when I got tired of the job and the pay and found one 65 miles away, in the closest metro area to where I live for my wife's (higher-paying) engineering job. The new job actually paid almost exactly 10% (plus benefits) more than the old one, as I suspected it would, as I was pretty underpaid at my old remote job.

Even commuting 65 miles each way every day, assuming fuel prices between $3.00-3.50/gal (which it is in my area, luckily), fuel costs only soak up about 40% of the gain, giving me a net gain of about (60% * 10% = ) 6%, plus benefits. Wear and tear on my car is pretty minimal, monetarily, as I drive a 22-year-old reliable sedan, on which I perform all regular maintenance and break-fix work myself, meaning that it's pretty cheap overall. Parking at work is free, and my commute is a straight shot on two-/four-lane highways, regularly averaging 80 mph with no real traffic to speak of. I listen to books on tape while I drive, and it makes it pretty stress-free.

In addition, I actually *love* my new job, and although I am authorized to telecommute one day a week, I usually do it more like one day a month, when I have commitments (doctor/dentist appointments, etc.) near home.

Overall, I think I've come out ahead trading a 10% raise for a real commute, and I wouldn't take a 10% pay cut to telecommute full-time. I don't think I'd even take a 5% cut. I wouldn't take a 10% pay *raise* (putting my right where I am now) to go back to my old job, as I felt under-valued as a team member, and didn't have any potential for upward mobility. Here, I have awesome benefits, a decent commute (which may be shortened significantly in the next year or two, if an expected transfer and relocation with my wife's job comes through), and I feel like a part of a dynamic team with significant potential for upward mobility and regular over-COLA raises.

But I recognize that I am probably a pretty unique case...

Slashdot Top Deals

PL/I -- "the fatal disease" -- belongs more to the problem set than to the solution set. -- Edsger W. Dijkstra, SIGPLAN Notices, Volume 17, Number 5