Slashdot Mirror


User: pstemari

pstemari's activity in the archive.

Stories
0
Comments
37
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 37

  1. Re:XMPP on Microsoft Embraces AMQP Open Middleware Standard · · Score: 1

    More to the point, "reliable" messaging is generally a reference (or euphemism, depending on your POV) for XA transaction support. Under XA transactions, you can combine things like database updates and queue get/puts so that they either all succeed or all fail atomically. Not so important for IM, but rather important when the database is an order cart and the message is debiting your bank account.

  2. Re:Too bad Java generics are completely useless on Java Generics and Collections · · Score: 1

    Generics are OK for the user, but when it comes to authoring code, they're much more error-prone that the C++ implementations. You essentially lose all type checking as soon as you use an array for the internal implementation, and wind up back to C++ anti-bloat hackery of using generics as a typecast wrapper around a non-generic implementation based on void* aka Object. At least in C++ you have the choice of marching down the cast wrapper path vs. writing a truly type-safe implementation that the compiler can fully optimize. Face it--this is basically syntax-sugar for the novice user, and suffers from all the usual issues with runtime dynamic type-checking.

  3. Re:Fine on New Mexico Might Declare Pluto a Planet · · Score: 1

    No, it doesn't. That definition not only includes a fair number of the larger asteroids, but the Moon to boot. Not only does the Moon's orbit never cross itself over the course of a year, it's actually concave AWAY from the Earth for half of its length.

  4. Lexmark is an IBM spinoff... on Lexmark Recalls 40,000 Laser Printers · · Score: 1

    ..and banks bleed blue. Banks have Lexmark printers because of the historic relationship with IBM.

  5. There is no "magenta" wavelength.... on RGB to become RGBCMY · · Score: 2, Insightful

    ...it's a mixture of red and blue from opposite ends of the spectrum. Cyan and yellow both depend on equally exciting both the green & blue and the red & green cones equally, but that can be accomplished by a swingle wavelength, unlike magenta.

  6. Re:The magic of studies on Embedded Systems Study Rebutted · · Score: 5, Insightful

    Nice troll, but it doesn't fly. There are in fact objective criteria for evalating experimental and statistical methodology, even though both coal companies and the "alternative medicine" crowd wish it weren't so.

  7. Re:How is this piracy? on DMCA Vs. The Sewing Underground · · Score: 1
    At best, the guy selling the patterns on the web is selling property that he doesn't own...at worst, he stole it.

    No and no. The Supreme Court ruled in a number of search and seizure cases that once you put something in the trash, it becomes abandoned property and is free for the taking.

    That's why the police can search your trash without a warrant. California vs Greenwood, 486 U.S. 35.

  8. I'll believe it when I see it... on Effective Vaccine For Malaria · · Score: 5, Informative
    Check out Robert Desowitz's The Malaria Capers for why taking a grain of salt with this announcement may be advised. The plasmodium parasite has multiple life stages that must be protected against to develop full immunity. Protection against one does not provide protection against the others.

    Also, calling anything based on the vaccinia virus "safe" in such a blanket fashion is vast overkill. Unlike most vaccines, vaccinia-based ones are live virus and cause severe complications if they get into the bloodstream--rather like the difference between cutaneous anthrax and pulmonary anthrax. Check out the CDC for just how nasty vaccinia can get if it escapes the vaccination site.

  9. Re:benefit of the doubt? on An Empirical Look at Software Patents · · Score: 4, Informative

    No, the blurb distinguishes services companies, which do work for hire, from software publishers.

    Pure software publishers obtain unusually few patents per $10m R&D, only about a quarter of the rate for the whole economy. (But software services firms, and especially IBM, patent significantly more).

    Remember, IBM Global Services is the largest work-for-hire outfit in the world.

    That result is more than slightly counter-intuitive, but there you have it.

  10. Re:Secret arrests on Former Intel Employee 'Disappeared' by U.S. · · Score: 1
    2: Far fewer informants

    How many Neighborhood Watch participants in your neighborhood? How many DARE programs asking children to report their parents for smoking pot?

  11. Re:Slashdot interview... on Analysis of SCO vs. IBM · · Score: 1
    If the non-Linux peers are steering the Linux developers at IBM, that's arguably violating the trade secrets IBM is contractually obliged to protect.

    Remember, trade secret != copyright != patent.

    If on the other hand, the BSDI decision can be used to show that the trade secrets were publically disclosed through no action of IBM, then SCO's argument goes up in smoke like dry ice on hot pavement. I'm not sure how that would turn out--the BSDI decision wasn't so much a win as a ceasefire declared after the plaintiffs discovered that they were at least equally if not more vulnerable to claims of copyright infringement.

  12. Re:No Official Reason? on Matrix Special Edition Cancelled · · Score: 1

    There's only so much shelf space available at the really critical locations--checkout lines, mid-aisle displays, etc. Access to that space is highly competitive and you can't really push two similar items at the same time.

  13. Re:Not the final solution on Debugging SMP Code with UML · · Score: 2, Interesting
    Depending on how the user-mode kernel is implemented, it may actually be better at reproducing SMP bugs than an actual SMP device. Typical SMP-only bugs are assumptions about cache operation that aren't seen on single processor machines with a single cache and synchronization issues on non-atomic memory access.

    An appropriately malicious emulation of SMP hardware can force these faults out much faster than they would appear in real life. One simple example would be cache emulation. IRL cache is limited and will be flushed automatically at some point. The emulation can treat cache as unlimited and only flush it on operations that guarantee cache flush. This will shake a lot of difficult-to-reproduce bugs out quickly.

  14. Re:Niggling correction on Rambus Destroyed Evidence In Anti-trust Trial · · Score: 1

    This is a civil case, not a criminal one. Preponderance of evidence suffices, and you can be compelled to provide evidence that adversely affects your case.

  15. Re:already done in the US? on Cat Organ Transplants · · Score: 1

    It's supportive therapy for cats with bad kidneys. It helps prevent the progression of kidney failure, which will otherwise progress exponentially.

  16. Re:already done in the US? on Cat Organ Transplants · · Score: 1
    Apparently cats don't have the genetic diversity of humans and only require minimal immuno-suppression. AFAIK (and believe me, I've checked), cats don't get dialysis. You can provide fluid therapy--i.e. 100-300ml of lactated Ringer's injected under the skin, and that does help quite a bit.

    My furball has been getting that daily for the last seven years, and he's quite happy. Even purrs while he's getting the fluids. Only side effects is that he sloshes for a while after the injection.

  17. Re:Polygraph == Non-science on Skepticism, Censorship And The Polygraph · · Score: 1

    Which is why refusal to take a polygraph is also not admissable.

  18. Re:Polygraph == Non-science on Skepticism, Censorship And The Polygraph · · Score: 1
    the polygraph is another example of officially endorsed psuedo-science.
    Not really. Even though elements of the counter-intelligence and law enforcement community think the polygraph is a useful tool, polygraph evidence is barred from every court of law in the US.
  19. Re:Eh? 3%? on The Future of the CD · · Score: 1
    The RIAA wouldn't, but it's the job of the reporter to ask those questions. Have we so quickly forgotten that the decline in CD revenues closely parallels the decline in the number of new releases?

    Oh, silly me. This was in the business section, home to as many fluff articles as the sports page.

  20. Re:Should you fear Google? on Should you Fear Google? · · Score: 1
    The funniest thing about the Google Watch site is that, despite their rantings about Page Rank, it comes up as #1 if you Google for it.

    Almost everything they complain about can be switched off, and is usually off by default:

  21. Re:Onwards and upwards on Final Mission for the Ariane-4 Successful · · Score: 1
    No, I meant the Ariane 5 looked like a Titan III knockoff.

    Depending on the exact configuration, the Ariane 4 can have up to four solid-fuel strap-ons. The AR40 has none, the AR42P has two solid-fuel strap-ons, and the AR44P has four. It can also use liquid fuel strap-ons, alone or in conjunction with solid fuel strap-ons, giving the AR42L (two liquid), AR44LP (two of each), and AR44L (four liquid). It's really an amazing amount of flexibility.

  22. Onwards and upwards on Final Mission for the Ariane-4 Successful · · Score: 4, Insightful
    Figures on the website indicate that the Ariane-5 will roughly double the payload to geosynch orbit. A rather nice feat for a program with a good history of success. Reliability should improve with more launches.

    Am I the only one who think it looks like a Titan-III knockoff, though?

  23. Re:de minimis on Negative Effects of Workplace Net Monitoring · · Score: 1
    Is taking a pen from work stealing?
    Heh. The feds used to have "Property of the US Government" embossed on the side of every pen.
  24. Re:linux should have non-exec stack by defualt on OpenBSD Gets Even More Secure · · Score: 1

    All of that doesn't help you when the desire to use a flat memory space limits you to a single data segment and a single code segment, set to begin at the same location.

  25. Re:linux should have non-exec stack by defualt on OpenBSD Gets Even More Secure · · Score: 1

    Old FORTRAN compilers didn't even use a stack. IBM mainframe compilers generated a static variable for each subroutine call, and swapped the value of R14 before making the call and restored it after return. That's why FORTRAN (and Cobol) didn't support recursion or truly local variables.