Hans leads police to Nina's body - Mandriva

This is a discussion on Hans leads police to Nina's body - Mandriva ; On Fri, 08 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article , Jennifer Coopersmith wrote: > Couple weeks ago, we had a storm go thru and a power surge into our >house. Took out my computer, 3 ports on ...

+ Reply to Thread
Page 6 of 14 FirstFirst ... 4 5 6 7 8 ... LastLast
Results 101 to 120 of 280

Thread: Hans leads police to Nina's body

  1. Re: [O/T] Hans leads police to Nina's body

    On Fri, 08 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , Jennifer Coopersmith wrote:

    > Couple weeks ago, we had a storm go thru and a power surge into our
    >house. Took out my computer, 3 ports on the 4 port router, and the
    >control board of our furnace/AC. All 3 are on 3 different circuit
    >breakers, and 2 of them on different sides of the breaker box.


    From your description, this sounds as if it were a lightning strike on
    the power lines, or a strike in very close proximity to your house.

    >Am thinking of getting a whole house surge protector. What are your
    >thoughts?


    Not really enough information. A whole hose surge protector may or
    may not be useful here, but is an expensive device. I've seen stories
    both ways. Results can vary significantly depending on how well it
    is installed, and where the surges are coming from.

    Whole house protections are meant to divert the damaging excess
    voltage - usually to ground. AC house wiring already should have
    one side grounded. The protector tries to limit the spike on the
    hot sides. Telephone are often floating (ungrounded), and protection
    often consists of devices that will arc over if the voltage gets to
    far away from ground (perhaps a hundred volts). Cable TV lines may
    have one side grounded, and have some kind of arc-over device. Both
    telephone and cable (and their broad-band networking equivalents)
    have this protection where the lines enter the house.

    A problem occurs when there is a lightning strike. The excess voltage
    is shunted to ground - but where is this "ground"? Many people assume
    the ground is that rod pounded into the earth generally close to the
    circuit breaker/fuse panel, or the cold water supply pipe where it
    enters the house. No, that is ground only if you are living in a
    salt water marsh. For everyone else, there is a measurable resistance
    between the ground rod and the sink/source of electrons. This results
    in the ground point being some voltage different from that you would
    find in a salt marsh (Ohms law says voltage equals resistance times
    current == volts equals ohms time amps).

    I had a microwave radio station located on an airfield. It consisted
    of a transmitter and antenna building, and a monitor located on a
    20 foot tall tower 120 feet away. Tower and building had lightning
    rods, connected to 25 foot long rods. There was a second set of rods
    at both sites, also 25 feet long that were the electrical ground for
    each site. The monitor signal was connected by buried cable in steel
    conduit to the building. There were surge protectors on everything,
    including at both ends of those wires between the monitor and
    transmitter building.

    Well, there was a lightning strike, about 300 feet away from the
    building - maybe 220 feet from the tower (there was a plane landing at
    the time, and we had multiple television cameras following it - the
    strike missed the plane by about 100 feet). The surge protectors on
    those wires between monitor and transmitter were toast, as was some of
    the electronics on the safe side of the protectors. Why? Evidence
    suggests that the lightning strike hit the earth, and the current
    between the strike point and the real ground generated a potential hill
    - "ground" was many volts above "ground" at the strike point, lesser
    number of volts above "ground" on the surface X feet radius away from
    the strike, and so on. This meant that the "ground" at the monitor was
    different than the "ground" at the transmitter building, and so on - to
    the extent that it exceeded the allowable voltage limits of the
    electrical components. The result was loss of the magic smoke in those
    components. So even though everything was grounded properly to the
    applicable standards, we lost - because of the difference in voltages.

    >Nothing else in the house was touched, nor any body else's on the
    >block. Could have lost some expensive toys.


    No doubt. Honestly, my suggestion would be to get a licensed
    electrician out to have a look at the grounding as it exists now. The
    damage to the furnace/AC controller suggests it was a surge that came
    through the power lines, and the components that died were just lucky
    (in the negative sense of the word). It may be that they were the most
    delicate, or there was a ground issue, or just bad karma. Would a
    whole house surge protector have made a difference? Possibly. Think of
    it as insurance, and look at the cost verses the value of damage a
    surge might cause, times the probability of such a surge occurring.
    This is a real case where "your mileage may vary". Weigh the facts
    carefully, as there are some serious smoke-and-mirror salesmen out there
    (I'm somewhat surprised a poster going by the nym that looks like 'x_xxx'
    hasn't posted
    here yet extolling his whole-house protection schemes with short leads
    to a real ground - makes you wonder how the Empire
    State Building or Sears Tower hits every year, and whose 80th floors are a bit more than ten feet
    from a "real" ground> hasn't been vacant for decades.)

    Old guy

  2. Re: [O/T] Hans leads police to Nina's body

    On Fri, 08 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    <6g3qrdFe2d6vU1@mid.individual.net>, bobbie sellers wrote:

    > You don't mention Mr.Franklin's invention for the prevention of
    >damage to houses, i.e. Lightening Rod? I am sure if you live in an
    >area prone to thunderstorms and lightening strikes you must have one
    >nearby?


    See my response to Jennifer Coopersmith up-thread. Lightning rods
    protect objects below a shallow cone centered on the top of the rod,
    but had (as an example) no effect on the lightning strike I mention
    even though there were two rods ~12 feet AGL and 300 feet away, and
    ~25 feet tall about 220 feet away. Also, like a kid jumping on a
    spring bed or trampoline, the strike _momentarily_ distorts the local
    ground reference - and that appears to have been what smoked some of
    the gear on that airfield. Good news for some, not for others.

    Old guy

  3. Re: [O/T] Hans leads police to Nina's body

    On Fri, 08 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , Adam wrote:

    Moe Trin wrote:

    >> Hard to say - this could be STRONGLY influenced by the software
    >> protocols used to transfer the data.

    >
    >I'm assuming (hoping) that when I plug in a USB external HD, the OS
    >will automatically recognize it and mount it, like it does with my USB
    >MP3 players which automatically become /mnt/disk (then /mnt/disk1 ...).
    >Then I could just use 'cp' to copy files. I hope.


    Yeah, that's probably going to be the case. It's when mounting them
    over a network that things become more complex, but that's mainly
    the setup. Once things are set up, it's as if they are mounted
    locally. They might be slightly slower, but that's about it.

    [compton ~]$ ls /net/
    hubble james.webb selene spitzer
    [compton ~]$

    NFS mounted stuff from four other computers.

    [sequence of pull-down menu items]

    >> otherwise, it obviously can't be done, and there is no use trying.

    >
    >Yep, I've been through a half hour of that without success, then edited
    >some config file with vim and solved the thing in under five minutes.


    In another group, someone is double clicking on some icon that is a
    tarball of a book from the LDP, and is wondering why it's coming up in
    gedit. Can't quite get the head around the command line 'zless
    /path/to/name.of.file.gz' for some reason.

    >> I've still got some cases that came with front-panel push-buttons,
    >> and fit plastic caps made from cut-down prescription drug containers
    >> so that you have to stick the finger into a short cylinder to
    >> actuate the switch.

    >
    >Until I cleared off part of my desk, sometimes the piles of papers
    >would press against the power switch, and that's 4.5" above the desk
    >top.


    First off, the computers are stacked along side the desk - the cats
    would not approve, and knock the papers out of the way. The desk
    belongs to Smokie - I'm merely allowed to set there when not serving
    or adoring her. There are RULES, you see.

    >>> The price range of desk chairs seems amazing, $25 to $400. I'll
    >>> look for something with armrests near the low end of that range.

    >>
    >> HIGHLY recommended. It makes a definite difference.

    >
    >High priority now, hopefully next week. There do seem to be models
    >under $100. There also seem to be models over $1000. Wow.


    A lot of it is what the public will tolerate. The last "decent" chair
    I bought (cloth seat/back/arm-rests, pneumatic height adjustment) was
    at Target and cost ~$50. My wife's chair is more "up-scale" and cost
    around $90 from Best Buy.

    >I figure I'll eventually get something along the lines of
    >http://www.apc.com/resource/include/...ase_sku=BE550R
    >which has 4 outlets with battery backup (e.g. tower, monitor), 4
    >surge-protected outlets without backup (e.g. printers), and phone line
    >surge protection. The power strip I'm using now would get disconnected.


    BackUPS ES-350 - yeah, that's a decent enough unit. Watch the run time
    at the load you may have. I've got some applications that take 3 minutes
    to shut down, and the computer shutdown after that takes another minute.
    That doesn't leave much time at ~3.3 minutes - especially given that I
    don't start an immediate shutdown on detecting a power fail. Also, over
    the life of the battery, you will loose some capacity, depending how
    hard you are cycling it.

    >This would give everything a common ground, and the UPS's power switch
    >would be a "master power switch" after the components have completed
    >their shutdown routines.


    Agree

    >That UPS's 45 dBA @ 1m sounds mildly annoying, though.


    That's actually about a third as noisy (5 dB) as the fans I've added
    to the boxes here. To me, it's background noise, that disappears
    unless I'm listening for it, or it changes.

    >There will be brief periods when there's a line-level audio connection
    >from my stereo system (which is on a separate circuit) to my computer,
    >while digitizing LPs or cassettes, but I don't think that should cause
    >any problems.


    As long as it's not during a period when the RIAA is around, or when
    there is a local lightning possibility. In theory, you could probably
    add an isolation transformer per channel, but that's almost certain
    to be overkill.

    >> Yup - in the 1980s, they had an outbreak of the Mediterranean Fruit
    >> Fly in central California

    >
    >They asked when we crossed the state line from Yuma, AZ. When we came
    >back from Mexico, they didn't ask. I dunno why.


    Dunno - did the Feds ask? There is a Mexican Fruit Fly but I think
    Mexico is fairly rigorous about control of that.

    Old guy

  4. Re: [O/T] Hans leads police to Nina's body

    On 2008-08-08, Jennifer Coopersmith wrote:
    > Moe Trin wrote:
    >
    >>>> I'd make do with a single surge protected power strip
    >>>> feeding both.
    >>> That's almost what I have now, everything off one surge-protected
    >>> power strip which is also a phone line surge protector. I hadn't
    >>> thought of plugging the UPS into the power strip, but that should
    >>> work, even though I'd no longer have one "master switch" for the whole
    >>> system.

    >>
    >> The only reason I was suggesting the single power strip is the surge
    >> protection - common grounds and all that. I wouldn't use the power
    >> switch on that strip normally. The output of my big UPS goes to a
    >> distribution strip, and plugged into that are small switched power
    >> strips (2-6 outlets - depending on where I found them) that I use
    >> for switching off power to the various computers/switch/monitor that
    >> are powered by that UPS. The printers (two networked, one on an A/B
    >> switch to the company computers) are powered from the common
    >> surge protectors in front of the UPSs.
    >>
    >> Old guy

    >
    > Hey Old Guy,
    > Couple weeks ago, we had a storm go thru and a power surge into our
    > house. Took out my computer, 3 ports on the 4 port router, and the
    > control board of our furnace/AC. All 3 are on 3 different circuit
    > breakers, and 2 of them on different sides of the breaker box. Am
    > thinking of getting a whole house surge protector. What are your
    > thoughts? Nothing else in the house was touched, nor any body else's on
    > the block. Could have lost some expensive toys.
    > (On that router, one port still works and so does the wireless
    > portion of it. Linksys WRT54G Now I have the WRT54G2)
    >
    > Jenn


    While a whole house surge protector won't solve all the
    world's problems, it's a pretty good idea.

    Lightning struck a power pole about 350 feet (~110m?) from
    my house a few years ago. I had a whole house surge
    protector wired into the entry panel and other surge
    supressors on outlets around the house. The only (known)
    electrical damage in my house was to a couple of line
    driver/receiver chips for the RS-422 link between the
    computers inside and the 32-zone sprinkler controller in the
    shed.

    --
    Robert Riches
    spamtrap42@verizon.net
    (Yes, that is one of my email addresses.)

  5. Re: [O/T] Hans leads police to Nina's body

    Moe Trin wrote:
    >> I'm assuming (hoping) that when I plug in a USB external HD, the OS
    >> will automatically recognize it and mount it, like it does with my USB
    >> MP3 players which automatically become /mnt/disk (then /mnt/disk1 ...).
    >> Then I could just use 'cp' to copy files. I hope.

    >
    > Yeah, that's probably going to be the case.


    I meant the MP3 players (which also work as flash drives) automatically
    became /media/disk, then /media/disk1, etc., not /mnt which is only
    non-removable storage. The instructions (of course) didn't mention
    Linux, but I figure just to be safe I do a 'sudo umount /media/*' before
    unplugging them. The Sansa 1GB MP3 player came with some programs or
    data files on it, purpose unknown, but I've left them on there because I
    don't know what will happen if I remove them. It does occasionally get
    plugged into Windows systems, maybe even a Mac someday. It also came
    with some songs I didn't like.

    > In another group, someone is double clicking on some icon that is a
    > tarball of a book from the LDP, and is wondering why it's coming up in
    > gedit. Can't quite get the head around the command line 'zless
    > /path/to/name.of.file.gz' for some reason.


    On the "it couldn't hurt" theory, I have clamav installed, and without
    too much trouble wrote a cron script to scan part of the system each
    night, so over a few days it scans everything. I can't imagine how to
    do that using GUIs.

    Back when I was into AL programming on the TRS-80 under LDOS/LS-DOS,
    there was an OS call to parse and evaluate command-line parameters. Is
    there anything comparable for bash scripts?

    >> Until I cleared off part of my desk, sometimes the piles of papers
    >> would press against the power switch, and that's 4.5" above the desk
    >> top.

    >
    > First off, the computers are stacked along side the desk - the cats
    > would not approve, and knock the papers out of the way. The desk
    > belongs to Smokie - I'm merely allowed to set there when not serving
    > or adoring her. There are RULES, you see.


    Once in the early '90s, Allegra (
    http://mysite.verizon.net/~adam707/Allegra.jpg ) must have felt I was
    spending too much time at the computer, so she came over and held down
    ctrl-c until CompuServe disconnected me.

    [desk chairs with armrests]
    > A lot of it is what the public will tolerate. The last "decent" chair
    > I bought (cloth seat/back/arm-rests, pneumatic height adjustment) was
    > at Target and cost ~$50. My wife's chair is more "up-scale" and cost
    > around $90 from Best Buy.


    I think I will just check the web pages (especially this week's sale
    flyers) for Staples, Office Depot, Best Buy, Target, and maybe Sears,
    pick out the best choice out of what's listed, then go to the store and
    buy it. (That leaves out KMart, WalMart, Circuit City, and Penney's,
    which also have local stores and may have chairs.) If I don't do
    something like that soon, I'll spend forever agonizing over it.

    > BackUPS ES-350 - yeah, that's a decent enough unit.


    I think I'd need something with more power than that. I used my
    Kill-A-Watt to measure the power consumption of each component (one of
    the main reasons I bought it), and one day I'll find my notes again. It
    was over 400 VA total for the pieces I'd want backup power for. It was
    interesting to see how the CRT monitor's power usage varied depending on
    the brightness of the image. From what I've heard, the APC products are
    good, so it would be simplest to just pick one of their products. Are
    there other brands that I should trust, or avoid?

    > Watch the run time at the load you may have.


    I'll have to measure poweroff time under heavy load.

    > I don't start an immediate shutdown on detecting a power fail.


    Makes sense to me. If I have a choice, around 15 seconds seems sensible
    for this area. It's interesting to see how my current box (Compaq) has
    a, not great, but better power supply than my previous (white box)
    system -- it's kept going through minor flickers that have reset digital
    clocks and would have shut the old one down.

    >> This would give everything a common ground,


    I can't seem to find it in the archives for this NG (I looked hard), but
    I remember asking you whether this building's ground rods for the power
    lines and phone lines (about 20 feet apart) were adequate, or I should
    install my own grounding rod, and I remember concluding that I didn't
    need to do anything about that, so you must have said the existing setup
    was good enough. I know I can't be the only renter here.

    >> That UPS's 45 dBA @ 1m sounds mildly annoying, though.

    >
    > That's actually about a third as noisy (5 dB) as the fans I've added
    > to the boxes here. To me, it's background noise, that disappears
    > unless I'm listening for it, or it changes.


    The laser printer's specs say 42 dBA @ 1M in standby (mainly the fan),
    and I find that mildly annoying. When I finally choose an external HD,
    noise will also be an important concern. For an apartment, this place
    is amazingly quiet 99% of the time, and I've gotten to love it. My last
    apartment was along the main westbound route through the city, with
    constant traffic, sirens, and other sounds. I just hope that whoever
    moves into the apartment next to mine will be quiet; it's supposed to be
    ready for occupancy next month.

    >> a line-level audio connection
    >> from my stereo system [...] while digitizing LPs or cassettes

    >
    > As long as it's not during a period when the RIAA is around


    I thought that digitizing my own LPs (or copying them to cassette, or
    making backups of my out-of-print CDs) for my own use was "fair use." I
    know that giving copies to others isn't allowed.

    >> They asked when we crossed the state line from Yuma, AZ. When we came
    >> back from Mexico, they didn't ask. I dunno why.

    >
    > Dunno - did the Feds ask? There is a Mexican Fruit Fly but I think
    > Mexico is fairly rigorous about control of that.


    Nope, Feds didn't ask when we re-entered California from Mexico, and we
    even had a large, open container of very nice strawberries right on the
    front seat between us that we'd bought while waiting for the border
    crossing.

    Adam

  6. Re: [O/T] Hans leads police to Nina's body

    On Sun, 10 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , Adam wrote:

    >Moe Trin wrote:


    >I meant the MP3 players (which also work as flash drives) automatically
    >became /media/disk, then /media/disk1, etc., not /mnt which is only
    >non-removable storage.


    That's an automounter of some kind.

    >The instructions (of course) didn't mention Linux, but I figure just to
    >be safe I do a 'sudo umount /media/*' before unplugging them.


    That's always a good idea - which is why I'm not using the automounter.
    I want to have some control on what gets written to the media and when.
    I've got the CD and DVD readers set to mount as a user, which means that
    user can mount/umount them.

    >The Sansa 1GB MP3 player came with some programs or data files on it,
    >purpose unknown, but I've left them on there because I don't know what
    >will happen if I remove them.


    What file system? (cat /etc/mtab) I'm guessing it's not one of the
    UNIX O/S file systems, which means that there is no real atime variable
    on the files.

    >On the "it couldn't hurt" theory, I have clamav installed, and without
    >too much trouble wrote a cron script to scan part of the system each
    >night, so over a few days it scans everything.


    Lack of DOS/windoze - which means an anti-virus application isn't going
    to do me much good. I do use a variation on 'tripwire'/'aide' to watch
    certain parts of the file system, but having to recreate the database
    every time I update the O/S is a bit of a pain.

    >I can't imagine how to do that using GUIs.


    Depends on how the GUI is designed. It might be one of the choices the
    GUI author decided would be useful.

    >Back when I was into AL programming on the TRS-80 under LDOS/LS-DOS,
    >there was an OS call to parse and evaluate command-line parameters. Is
    >there anything comparable for bash scripts?


    I'm not sure what you are asking here. The command line parameters are
    certainly available within the shell - $* is all, $0 is the name used
    to call the script/application, $1 ... $n being the individual
    parameters ($# is the number of parameters - 'man bash' and look at the
    PARAMETERS section).

    >Once in the early '90s, Allegra (
    >http://mysite.verizon.net/~adam707/Allegra.jpg ) must have felt I was
    >spending too much time at the computer, so she came over and held down
    >ctrl-c until CompuServe disconnected me.


    Another advantage of the sliding under-desk keyboard tray. Smokie has
    a habit of leaping up onto the desk, and as the key tray is somewhat
    lower and sticks out, she hits that pretty often. She _tends_ to avoid
    the keyboard, but...

    >I think I will just check the web pages (especially this week's sale
    >flyers) for Staples, Office Depot, Best Buy, Target, and maybe Sears,
    >pick out the best choice out of what's listed, then go to the store and
    >buy it. (That leaves out KMart, WalMart, Circuit City, and Penney's,
    >which also have local stores and may have chairs.) If I don't do
    >something like that soon, I'll spend forever agonizing over it.


    Yeah, you definitely do want to look at what's there, and try it if
    possible. But I do understand the procrastination factors too.

    >> BackUPS ES-350 - yeah, that's a decent enough unit.

    >
    >I think I'd need something with more power than that.


    Obviously, they do build larger ones in the BackUPS series.

    >I used my Kill-A-Watt to measure the power consumption of each
    >component (one of the main reasons I bought it), and one day I'll find
    >my notes again. It was over 400 VA total for the pieces I'd want
    >backup power for.


    I like to have a bit of reserve power - as I know I'll be adding
    something that might push it over the edge, and my meter is only quoted
    as 3% as manufactured (which is the last time it was calibrated).

    >It was interesting to see how the CRT monitor's power usage varied
    >depending on the brightness of the image.


    It's an expensive means of converting volt-amps into lux, but...

    >From what I've heard, the APC products are good, so it would be
    >simplest to just pick one of their products. Are there other brands
    >that I should trust, or avoid?


    Sola and Tripp Lite have some good models - we're using one of the
    latter on my wife's company system (no idea which model), but the
    general preference is for APC.

    >> I don't start an immediate shutdown on detecting a power fail.

    >
    >Makes sense to me. If I have a choice, around 15 seconds seems
    >sensible for this area.


    We get momentary outages - a second or so - with some regularity.
    The ones I have to be concerned with tend to be much longer. I have
    the systems set to start the shutdown a minute after the AC line goes
    away, and that seems to be acceptable. If it's over a minute, it's
    _probably_ going to be out longer than the UPS can carry it.

    >It's interesting to see how my current box (Compaq) has a, not great,
    >but better power supply than my previous (white box) system -- it's
    >kept going through minor flickers that have reset digital clocks and
    >would have shut the old one down.


    That usually means bigger caps on the primary DC rectifier.

    >I can't seem to find it in the archives for this NG (I looked hard),
    >but I remember asking you whether this building's ground rods for the
    >power lines and phone lines (about 20 feet apart) were adequate,


    a.o.l.mandrake

    >or I should install my own grounding rod, and I remember concluding
    >that I didn't need to do anything about that, so you must have said
    >the existing setup was good enough. I know I can't be the only renter
    >here.


    I suspect we discussed the common surge protector concept. The problem
    with separate grounds is if there is a strike close by (say less than
    300 feet) or a direct hit on one of the services. The nearby strike
    causes a differential in the local ground voltage as a function of
    distance from the strike, while the direct hit could produce large
    currents in the ground associated with that service. Because grounds
    really aren't _absolute_ ground, and have a finite resistance to that
    magical place, fault currents with raise the local ground voltage as a
    function of resistance and current. Your adding a separate common
    ground point clamps (For Strange Values Of 'clamp') the voltages on the
    various protected leads relative to each other.

    >>> That UPS's 45 dBA @ 1m sounds mildly annoying, though.

    >>
    >> That's actually about a third as noisy (5 dB) as the fans I've added
    >> to the boxes here. To me, it's background noise, that disappears
    >> unless I'm listening for it, or it changes.

    >
    >The laser printer's specs say 42 dBA @ 1M in standby (mainly the fan),
    >and I find that mildly annoying.


    I've got enough other noise running about (fans, humidifier, radio,
    etc.) that noise at that level is difficult to perceive. I notice it
    when there is a change - but it's pretty well filtered out by the
    brain filters.

    >When I finally choose an external HD, noise will also be an important
    >concern. For an apartment, this place is amazingly quiet 99% of the
    >time, and I've gotten to love it. My last apartment was along the
    >main westbound route through the city, with constant traffic, sirens,
    >and other sounds. I just hope that whoever moves into the apartment
    >next to mine will be quiet; it's supposed to be ready for occupancy
    >next month.


    The houses here are fairly well insulated, so external noise is not
    much of a factor. I'm directly under one of the departure corridors
    out of Phoenix Sky Harbor, but they're above 5000 AGL at this point.
    Last week, while in the pool (at 8PM), I noted that the major source
    of noise seems to be the air-conditioners - and the wild life (coyotes
    howling over a half mile away). There is a divided dual carriage way
    about a third of a mile away, but the wind has to be in the right
    direction to hear much traffic (sirens excepted).

    >> As long as it's not during a period when the RIAA is around

    >
    >I thought that digitizing my own LPs (or copying them to cassette, or
    >making backups of my out-of-print CDs) for my own use was "fair use."


    IANAL - I know that some of my LPs have "thou shalt not" notices on
    the jackets.

    >> Dunno - did the Feds ask? There is a Mexican Fruit Fly but I think
    >> Mexico is fairly rigorous about control of that.

    >
    >Nope, Feds didn't ask when we re-entered California from Mexico, and
    >we even had a large, open container of very nice strawberries right on
    >the front seat between us that we'd bought while waiting for the
    >border crossing.


    I'm not sure what the California regulations actually say - I know that
    the Feds will nail you for certain uncooked meats. I suspect the state
    rules are aimed at stuff that is not for immediate consumption by the
    person carrying them.

    Old guy

  7. Re: [O/T] Hans leads police to Nina's body

    Moe Trin wrote:
    > I've got the CD and DVD readers set to mount as a user, which means that
    > user can mount/umount them.


    I never looked into it, but Mandriva automounts data DVDs but not data
    CDs on my system. Gotta look into packet writing one of these days too.

    >> The Sansa 1GB MP3 player came with some programs or data files on it,
    >> purpose unknown, but I've left them on there because I don't know what
    >> will happen if I remove them.


    BTW those files are under 9M, so it's not a big sacrifice. At least
    some of them are data it creates and uses when playing MP3s.

    > What file system? (cat /etc/mtab) I'm guessing it's not one of the
    > UNIX O/S file systems, which means that there is no real atime variable
    > on the files.


    Both are 'vfat', not terribly surprising since that's RW by both Windows
    and Macs as is. Of course that means a lot of the attributes get lost
    when I use it to transfer files to/from Linux.

    >> On the "it couldn't hurt" theory, I have clamav installed, and without
    >> too much trouble wrote a cron script to scan part of the system each
    >> night, so over a few days it scans everything.

    >
    > Lack of DOS/windoze - which means an anti-virus application isn't going
    > to do me much good.


    And it can't find viruses within a VM either. Like I said, the "it
    couldn't hurt" theory.

    > I do use a variation on 'tripwire'/'aide' to watch
    > certain parts of the file system, but having to recreate the database
    > every time I update the O/S is a bit of a pain.


    I had 'aide' set to compare and update everything every night, until I
    realized that I almost never looked at the output, and when I did I
    couldn't tell usual from unusual file changes.

    > The command line parameters are
    > certainly available within the shell

    [snip]

    Well, I seem to have something like:

    { start of excerpt from bash script }

    {start of loop}
    case "$1" in
    -c)
    CFLAG=1
    ;;
    -p)
    PFLAG=1
    ;;
    -v)
    VERBOSITY=$2
    shift
    ;;
    *)
    echo "Unknown parameter $1"
    ;;
    esac
    shift
    done

    { end of excerpt from bash script }

    in any script with any parameters, and I basically have to rewrite the
    same parameter parsing code in each script with minor changes depending
    on what parameters are expected. Is there any way to avoid essentially
    recoding the same thing? Or is there an all-purpose routine out there
    that I can just paste into my scripts? (I realize this may be a stupid
    question, but I'm still new at writing bash scripts and hope the
    veterans will forgive me.)

    >> I think I will just check the web pages (especially this week's sale
    >> flyers) for Staples, Office Depot, Best Buy, Target, and maybe Sears,
    >> pick out the best choice out of what's listed

    [snip]
    >
    > Yeah, you definitely do want to look at what's there, and try it if
    > possible. But I do understand the procrastination factors too.


    Staples had some good sales this week, so with all the discounts it came
    to under $45. It's actually a lot like my previous chair except that it
    has arms. I can feel the difference already.

    >> I used my Kill-A-Watt to measure the power consumption of each
    >> component (one of the main reasons I bought it), and one day I'll find
    >> my notes again. It was over 400 VA total for the pieces I'd want
    >> backup power for.

    >
    > I like to have a bit of reserve power - as I know I'll be adding
    > something that might push it over the edge, and my meter is only quoted
    > as 3% as manufactured (which is the last time it was calibrated).


    I think trying to calibrate my Kill-A-Watt would be more trouble than
    it's worth, so I'll just add up everything, then add maybe 50% more to
    allow for inaccurate measurements and future expansion. It doesn't cost
    much more to get the next bigger UPS.

    > We get momentary outages - a second or so - with some regularity.
    > The ones I have to be concerned with tend to be much longer. I have
    > the systems set to start the shutdown a minute after the AC line goes
    > away, and that seems to be acceptable. If it's over a minute, it's
    > _probably_ going to be out longer than the UPS can carry it.


    Makes sense to me. In the 2.5 years I've lived here, I can only recall
    one outage that was longer than a few seconds, and there haven't been
    many short ones either.

    >> It's interesting to see how my current box (Compaq) has a, not great,
    >> but better power supply than my previous (white box) system -- it's
    >> kept going through minor flickers that have reset digital clocks and
    >> would have shut the old one down.

    >
    > That usually means bigger caps on the primary DC rectifier.


    That would do it. Of course those cost a few cents more, and the power
    supply is where they usually cut costs.

    >> I can't seem to find it in the archives for this NG (I looked hard),
    >> but I remember asking you whether this building's ground rods for the
    >> power lines and phone lines (about 20 feet apart) were adequate,

    >
    > a.o.l.mandrake


    Thanks, found it! I was using Google Groups to search in
    "alt.os.linux.mandr*" which apparently doesn't understand wildcards.

    > Your adding a separate common
    > ground point clamps (For Strange Values Of 'clamp') the voltages on the
    > various protected leads relative to each other.


    Sounds like my surge-protected power strip, or the UPS that will someday
    replace it, will be sufficient. Always nice to find something that
    doesn't need any more work!

    >>>> That UPS's 45 dBA @ 1m sounds mildly annoying, though.


    I don't understand why a UPS would make any noise at all. Isn't it just
    some electronics and a battery?

    Adam

  8. Re: [O/T] Hans leads police to Nina's body

    On Tue, 12 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , Adam wrote:

    >Moe Trin wrote:


    >> I've got the CD and DVD readers set to mount as a user, which means
    >> that user can mount/umount them.

    >
    >I never looked into it, but Mandriva automounts data DVDs but not data
    >CDs on my system.


    There are a number of automounters, and each one does things in a
    different manner.

    >Gotta look into packet writing one of these days too.


    ??? You mean like netcat?

    [compton ~]$ whatis nc
    nc (1) - TCP/IP swiss army knife
    [compton ~]$

    >> Lack of DOS/windoze - which means an anti-virus application isn't
    >> going to do me much good.

    >
    >And it can't find viruses within a VM either. Like I said, the "it
    >couldn't hurt" theory.


    I'd rather the "it might help" theory.

    >> I do use a variation on 'tripwire'/'aide' to watch certain parts of
    >> the file system, but having to recreate the database every time I
    >> update the O/S is a bit of a pain.

    >
    >I had 'aide' set to compare and update everything every night, until I
    >realized that I almost never looked at the output, and when I did I
    >couldn't tell usual from unusual file changes.


    I don't automatically update the system software, so it's RELATIVELY
    easy to keep track of what got changed (manually) there. My home
    directories are mounted from (local) file servers, and I try to
    minimize risk by being somewhat careful. Long ago, I started using
    unprivileged user accounts when visiting prospectively dodgy sites
    (they only have write permission to two very specific directories,
    neither of which is important).

    >> The command line parameters are certainly available within the shell

    >[snip]
    >
    >Well, I seem to have something like:
    >
    >{ start of excerpt from bash script }


    [snip]

    OK - standard case statement with default

    >in any script with any parameters, and I basically have to rewrite the
    >same parameter parsing code in each script with minor changes depending
    >on what parameters are expected. Is there any way to avoid essentially
    >recoding the same thing? Or is there an all-purpose routine out there
    >that I can just paste into my scripts? (I realize this may be a stupid
    >question, but I'm still new at writing bash scripts and hope the
    >veterans will forgive me.)


    Normally, when I write a script where there will be more than one
    argument, I start the script by listing in a comment, the variables
    and what they are, then checking the number of variables and barfing
    on a Usage: if the number isn't what I'd expect:

    [compton ~]$ head -14 bin/monthly.news.stats
    #!/bin/bash
    # $1 month, optional $2 file_to_process
    if [ $# -eq 1 ] ; then
    FILE="/var/spool/slrnpull/log"
    elif [ $# -ne 2 ] ; then
    echo "Usage: $0 two-digit-month /path/to/logfile"
    exit 1
    else
    FILE=$2
    fi
    TMPFILE=`mktemp /tmp/monthly.stats.XXXXXX`
    FILENAME=`basename $FILE`
    MONTH=$1
    rm /tmp/newsstats.$FILENAME.$MONTH 2>/dev/null
    [compton ~]$

    Standard bang-spat, comment, test number of variables supplied ($#),
    if only one, assume a filename, else if not two, Usage and bail, else
    assign second variable as filename. Create a tempfile name, remove (if
    it exists) what will be used as the output file, and so on.

    I guess I've been doing this for so long, I just don't think about it.
    It's just a case of "what do you have, what do you want, how do you get
    from A to B, and make notes/comments of this isn't a one-time deal.

    Writing scripts... The standard training track I tend to recommend is
    the 'Bash-Prog-Intro-HOWTO', followed by two guides from the LDP
    (http://tldp.org/guides.html) - the 'Bash Guide for Beginners' and the
    'Advanced Bash-Scripting Guide'. Another learning tool is looking at
    /etc/inittab, and following the scripts it uses to boot the system. You
    will find the man pages helpful there (man inittab to see what the file
    means, then heavy into man bash to understand how they did that). The
    reason I recommend trying to follow the boot scripts is that they are
    written by expert shell scripters, and they are FLAUNTING IT like it's
    going out of style.

    >Staples had some good sales this week, so with all the discounts it came
    >to under $45. It's actually a lot like my previous chair except that it
    >has arms. I can feel the difference already.


    Sounds good - Having a decent chair, with the keyboard and monitor at
    the right level makes a HUGE different in comfort. I average over nine
    hours a day at the computer, plus maybe another ten hours on the weekend
    and I don't think I could continue to do this if things weren't right.

    >I think trying to calibrate my Kill-A-Watt would be more trouble than
    >it's worth, so I'll just add up everything, then add maybe 50% more to
    >allow for inaccurate measurements and future expansion. It doesn't
    >cost much more to get the next bigger UPS.


    I can't remember when you bought that, but out of box it's supposed to
    be good to under a percent. How _long_ it retains that accuracy is
    another question. You could check it for *crude* accuracy by looking
    at a 100 Watt incandescent lamp (look for 10 percent accuracy there).
    Most appliances, whether electronic or something dumb like a fan or
    hair dryer, are listing Optimistic Watts (average, or worst case -
    which ever makes them look best), and you'll need the whole shaker of
    salt, not just a grain or two.

    50% might be a bit of an overkill unless it's relatively cheap.

    >> We get momentary outages - a second or so - with some regularity.
    >> The ones I have to be concerned with tend to be much longer.


    >Makes sense to me. In the 2.5 years I've lived here, I can only
    >recall one outage that was longer than a few seconds, and there
    >haven't been many short ones either.


    Most of the outages I see are either a system failure somewhere (and
    usually semi-automatically fixed in a second or two) or damage to the
    local distribution system - a drunk misses the curve (or straight) and
    drives through a pole or transformer (most of our utilities are under-
    ground, but the street transformers are on the side of the road, and
    inviting targets). Damage tends to take several hours to correct. Two
    or three weeks ago, we had a series of thunderstorms, and a micro-burst
    (intense gust) blew down about a half mile of power lines about ten
    miles from here. That didn't effect me, but it took the power company
    nearly two days to get everyone back on line. That happens two-three
    times a year, even with steel power poles.

    >> That usually means bigger caps on the primary DC rectifier.

    >
    >That would do it. Of course those cost a few cents more, and the
    >power supply is where they usually cut costs.


    It's also a physical size issue, but basically yes. They are buying
    in quantity 10,000 lots, but as a *CRUDE* rule of thumb, the retail
    price of the end product is about ten times the cost of the materials.
    Save a buck on the price of the caps, and that's ten bucks on the sale
    price before taxes.

    >> a.o.l.mandrake

    >
    >Thanks, found it! I was using Google Groups to search in
    >"alt.os.linux.mandr*" which apparently doesn't understand wildcards.


    Advanced search? Yeah, I've seen that problem, but they seem to be
    getting worse over time.

    >> Your adding a separate common ground point clamps (For Strange Values
    >> Of 'clamp') the voltages on the various protected leads relative to
    >> each other.

    >
    >Sounds like my surge-protected power strip, or the UPS that will someday
    >replace it, will be sufficient.


    For that goal, yes. I was at a trade show some years ago, and someone
    was flogging some kind of surge protection. They had a demo set up with
    an isolation transformer to protect the world, an impulse generator, and
    various surge protectors along with a 'scope to display what was going
    on. The klown was a marketeer, and so was dramatizing things a bit, but
    the 'scope traces were pretty convincing. The impulse generator wasn't
    that hefty (couple of joules, narrow pulses, 60Hz), but seeing kilovolt
    spikes attenuated to volts was impressive. He also demo'ed the fact
    that ground resistance was important (even with a couple ohm resistor
    in the ground lead, you could see the current spike).

    >>>>> That UPS's 45 dBA @ 1m sounds mildly annoying, though.

    >
    >I don't understand why a UPS would make any noise at all. Isn't it
    >just some electronics and a battery?


    There are transformers - the laminations are going to sing some. The
    large (2500 VA) UPS at home has a built in fan, though most smaller
    units lack this.

    Old guy

  9. Re: [O/T] Hans leads police to Nina's body

    Moe Trin wrote:
    >> Gotta look into packet writing one of these days too.

    >
    > ??? You mean like netcat?


    Nope, like using UDF format on RW media.
    /usr/share/doc/kernel-doc/filesystems/udf.txt has enough answers to get
    me started.

    > Long ago, I started using
    > unprivileged user accounts when visiting prospectively dodgy sites
    > (they only have write permission to two very specific directories,
    > neither of which is important).


    Good idea -- I should do the same.

    > Normally, when I write a script where there will be more than one
    > argument, I start the script by listing in a comment, the variables
    > and what they are, then checking the number of variables and barfing
    > on a Usage: if the number isn't what I'd expect:


    But what about options? Is the "case $1 in" loop what's commonly used?
    It took me a while to realize that for a return code, 0 is OK and
    non-zero is error, but for flag variables in scripts, false should be 0
    and true is non-zero... I think I got that right.

    > I guess I've been doing this for so long, I just don't think about it.
    > It's just a case of "what do you have, what do you want, how do you get
    > from A to B, and make notes/comments of this isn't a one-time deal.


    Okay, so there's really no way to avoid writing it once, but I could
    reuse it with minor modifications in other scripts.

    > Writing scripts... The standard training track I tend to recommend is
    > the 'Bash-Prog-Intro-HOWTO', followed by two guides from the LDP
    > (http://tldp.org/guides.html) - the 'Bash Guide for Beginners' and the
    > 'Advanced Bash-Scripting Guide'.


    Thanks for the references. Most of my scripting questions are either
    about syntax, or which (if any) tool will accomplish something.

    >> It's actually a lot like my previous chair except that it
    >> has arms. I can feel the difference already.

    >
    > Sounds good - Having a decent chair, with the keyboard and monitor at
    > the right level makes a HUGE different in comfort.


    I think it will take a few days and a little trial-and-error to get the
    best settings for seat height and armrest height (I'm average height),
    but your advice was excellent. In fact, I'm going to suggest ergonomics
    as a future LUG presentation.

    >> I was using Google Groups to search in
    >> "alt.os.linux.mandr*" which apparently doesn't understand wildcards.

    >
    > Advanced search? Yeah, I've seen that problem, but they seem to be
    > getting worse over time.


    Yep, Google Groups Advanced Search. At least there's only 2 NGs
    involved in that case. Trying to search through multiple Linux NGs
    simultaneously sounds awful.

    On the positive side, generally, if you use Google web search and one of
    the keywords is "Mandriva," it also searches for "Mandrake."

    [Kill-A-Watt power meter]
    > I can't remember when you bought that, but out of box it's supposed to
    > be good to under a percent. How _long_ it retains that accuracy is
    > another question. You could check it for *crude* accuracy by looking
    > at a 100 Watt incandescent lamp (look for 10 percent accuracy there).


    Okay, for a 60W GE incandescent bulb it says 57 W & 57 VA. For a
    three-way 50/100/150W Sylvania incandescent bulb, it reports 46 W & 46
    VA, 95 W & 94 VA, and 143 W & 142 VA. (I don't understand how watts
    could be greater than volt-amps.) Either it's reading 5-8% low, or the
    bulb manufacturers are chiseling.

    >> I think trying to calibrate my Kill-A-Watt would be more trouble than
    >> it's worth, so I'll just add up everything, then add maybe 50% more to
    >> allow for inaccurate measurements and future expansion. It doesn't
    >> cost much more to get the next bigger UPS.


    > 50% might be a bit of an overkill unless it's relatively cheap.


    Just as an example, if I decide to go with APC's ES series, the choices
    will be 330W/550VA, 390W/650VA, and 450W/750VA, $63, $83, and $100 at
    Newegg. Not really that much difference. Under the same load and AC
    power, would there be much difference in power consumed and noise
    produced with a larger UPS?

    >>>>>> That UPS's 45 dBA @ 1m sounds mildly annoying, though.

    >> I don't understand why a UPS would make any noise at all. Isn't it
    >> just some electronics and a battery?

    >
    > There are transformers - the laminations are going to sing some.


    45 dB sounds awfully loud for a home transformer. OTOH I haven't spent
    much time around transformers bigger than a few watts.

    > Two
    > or three weeks ago, we had a series of thunderstorms, and a micro-burst
    > (intense gust) blew down about a half mile of power lines about ten
    > miles from here. That didn't effect me, but it took the power company
    > nearly two days to get everyone back on line. That happens two-three
    > times a year, even with steel power poles.


    I haven't yet been in an outage that lasted past the following morning.
    However, longer ones seem to happen more often in the more remote
    parts of the county.

    >> the power supply is where they usually cut costs.

    >
    > It's also a physical size issue, but basically yes. They are buying
    > in quantity 10,000 lots, but as a *CRUDE* rule of thumb, the retail
    > price of the end product is about ten times the cost of the materials.
    > Save a buck on the price of the caps, and that's ten bucks on the sale
    > price before taxes.


    Do that on a few more parts, and you've got a model that sells for $100
    less than the competition where most people can't tell the difference.
    I didn't know about that rule, but I did find out that hard goods seem
    to sell for about double the store's cost, less on expensive items, more
    on cheap items.

    Adam

  10. Re: [O/T] Hans leads police to Nina's body

    In article
    ,
    Jim Beard in alt.os.linux.mandriva wrote:

    >Jennifer Coopersmith wrote:
    >> Hey Old Guy,
    >> thinking of getting a whole house surge protector. What are your
    >> thoughts? Nothing else in the house was touched, nor any body else's on
    >> the block. Could have lost some expensive toys.

    >
    >Different old guy. Maybe not as old nor as expert. That said,
    >A whole house surge protector/grounding is a good idea. It does
    >need to be done right, or you gain nothing.
    >
    >First, you need to ground your telephone line as well as the
    >electrical wiring.




    That may be possible in some countries, but it's bloody dangerous (and
    illegal) to do that in most European countries. Here, the phone line
    has a single earth point at the positive pole of the battery in the
    exchange. AC signals at the receiver are isolated by a transformer.
    All of the telephone wiring forms an equipotential zone for all
    subscribers on that exchange. The power grid is earthed at the star
    point /neutral of the distribution transformer (usually 750-1000 kVA)
    and taken to the consumer in a five or three conductor cable. Any
    metallic service pipes are bonded to the earth conductor at every
    consumer's incoming terminals. Thus everyone is in another
    equipotential zone, probably at a slightly different voltage to true
    earth. Then someone comes along with a JCB and digs up a power cable.
    You can spot when this happens because JCBs use a special paint - it
    changes colour from yellow to black. No problems, the earth fault
    current gets back to the transformer, raising the earth voltage by a
    few tens of volts, but the phone line remains unaffected - its in a
    different equipotential zone. Earth the phone line at a subscriber's
    premises, guess where the current flows. If you're lucky it will just
    melt the incoming phone cable.

    Sometimes you can see this when a multipair screened phone cable, as
    thick as your wrist, is carried a few miles on poles. Somehow the
    screen gets earthed at both ends. Nothing really bad happens until an
    earth fault on the leccy transmission grid. Cable melts.

    Alternatively, if only one subscriber earths the phone line, nothing
    really bad happens until a fault. If two subscribers fed from the
    same exchange but different distribution transformers earth the phone
    line, then a current will flow according to the difference in
    potential of the two equipotential zones.

    We suffered a few years back when there were peecee on most desks and
    Laplink connexions abounded. (Our guys didn't like sneakernet, they
    resented walking around the plot.) Some cables went out of a window
    in the main office block to temporary pre-fabricated offices
    alongside. What they didn't know was that the temporary buildings
    were powered from a transformer in the factory, a different
    transformer to the offices. We fell over this one some years earlier.
    Someone strung a 80 mA TTY cable across the works about half a mile
    to the KDF9s. This was done when the works were on their 14 day
    holiday shut down. (It worked well, no more mile walks in the rain to
    the terminal room.) The fire started at around 07:45 the day the
    works people returned, switched on their machinery, and caused a 20V
    difference between distribution transformer star points/earths. We
    had 12 different distribution transformers on the plot. Never earth
    the screen of a serial link at both ends. It can happen to you at
    home, transformer service areas do have boundaries. Probably OK to
    connect to the guy next door, but not to the guy on an adjacent street
    at the bottom of the garden.

    >Second, the grounding has to be an
    >effective ground, which generally means a copper shaft driven
    >deep into the ground. How deep it has to be depends on the
    >soil and the amount of moisture in it.
    >


    Always drive them in pairs, several feet apart. Fit bolted test links
    in each nest so that you can measure the earth impedance (it can
    change over time). Some amateur radio handbooks have tables of
    number of rods needed for different impedances and soils. You don't
    get much of an earth from one rod.

    >There are ins and outs to doing it right that I cannot address.


    Be safe, don't earth the phone line!

    >If anyone tells you it is simple, he is either lying or he has
    >done it so many times he could plan it in his sleep and install
    >it while sleepwalking. For a few guys like that, it is simple.
    >
    >A lightning rod is likewise a good idea.
    >
    >EMP is real, and in extremely rare circumstances can do things
    >that nothing can protect against. But the wierd forms of ball
    >lightening and other intractable problems (at least intractable
    >using today's knowledge and technology) are extremely extremely
    >rare. Ignore the EMP problem.
    >
    >Bottom line: The cheapest whole-house protection I have ever
    >seen offered was by my local electric company that mounted a
    >campaign to do thousands of houses over the summer. They charged
    >about $100, and that without putting in a lightening rod. Just
    >protection for phone lines and electric lines. You could easily
    >spend several times that to get someone competent to do the full job.
    >


    Please folks, the stuff is called "lightning".



    Regards,

    David P.


  11. Re: [O/T] Hans leads police to Nina's body

    On Fri, 15 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , David Powell wrote:

    >Jim Beard wrote:


    >>First, you need to ground your telephone line as well as the
    >>electrical wiring.

    >
    >


    I'd tend to agree. Bolting either side of the phone line to ground
    isn't the best idea. What _IS_ desired is to use devices - whether they
    be spark gaps, MOVs, TranZorbs, or Surgistors - to clamp (limit) the
    difference in the voltage between said lines and local references.

    >That may be possible in some countries, but it's bloody dangerous
    >(and illegal) to do that in most European countries. Here, the phone
    >line has a single earth point at the positive pole of the battery in
    >the exchange.


    I'm not a telephone expert, but my understanding is that the lines are
    not hard grounded. Certainly, tests on my home phones show the lines
    are not well grounded.

    >AC signals at the receiver are isolated by a transformer. All of the
    >telephone wiring forms an equipotential zone for all subscribers on
    >that exchange.


    I know what you are saying, but that doesn't appear to be the case here.

    >The power grid is earthed at the star point /neutral of the
    >distribution transformer (usually 750-1000 kVA) and taken to the
    >consumer in a five or three conductor cable. Any metallic service
    >pipes are bonded to the earth conductor at every consumer's incoming
    >terminals.


    Very different from the US. First, the supply transformer is local,
    and because of I*R drop in the wires, is unlikely to be more than a
    very few hundred feet - commonly less than 200 feet - from the
    transformer to the customer. There is a _required_ ground at each
    customer's demarc, and at the transformer. The ground is a safety
    feature, and isn't meant to be carrying currents except during
    fault conditions. Most residential distributions are a split
    single phase (center tapped winding), providing 120 +/- 6 VAC
    from each hot wire to the neutral. Between the (normal) two hot
    wires, you get 240 +/- 12 VAC unless you are actually getting two
    of three phases (208 +/- 10 VAC).

    >Earth the phone line at a subscriber's premises, guess where the
    >current flows. If you're lucky it will just melt the incoming phone
    >cable.


    Highly unlikely - the telephone is an energy limited device (line
    impedances). The normal wires used depend on distances when they
    were originally installed, but vary from #28 AWG (0.0126" dia, 64.9
    Ohms/1000 feet) to #19 AWG (0.0359" dia, 8.05 Ohms/1000 feet).

    >Sometimes you can see this when a multipair screened phone cable, as
    >thick as your wrist, is carried a few miles on poles. Somehow the
    >screen gets earthed at both ends. Nothing really bad happens until an
    >earth fault on the leccy transmission grid. Cable melts.


    Yes, that could happen - but we don't use screened cable - it's all
    just twisted pairs. There may be a steel messenger wire molded into
    the cable jacket, but these tend not to be grounded.

    >Some cables went out of a window in the main office block to
    >temporary pre-fabricated offices alongside. What they didn't know was
    >that the temporary buildings were powered from a transformer in the
    >factory, a different transformer to the offices. We fell over this
    >one some years earlier. Someone strung a 80 mA TTY cable across the
    >works about half a mile to the KDF9s. This was done when the works
    >were on their 14 day holiday shut down. (It worked well, no more
    >mile walks in the rain to the terminal room.) The fire started at
    >around 07:45 the day the works people returned, switched on their
    >machinery, and caused a 20V difference between distribution
    >transformer star points/earths.


    This indicates a power problem, as there should not be current flowing
    in the ground leads. This also indicates a severe problem with idiots
    allowed to run wiring in an unsafe manner.

    >We had 12 different distribution transformers on the plot.


    So? I'm sorry, but I've got three three phase transformers in the
    building I work in. There are telephone lines in the building coming
    from one telephone room... in a different building. There are five
    buildings on this site, each with multiple three phase distribution
    points. We don't have grounding problems simply because the grounds
    are not carrying currents. That's what the neutral line is for.

    >Never earth the screen of a serial link at both ends.


    Agreed

    >It can happen to you at home, transformer service areas do have
    >boundaries. Probably OK to connect to the guy next door, but not to
    >the guy on an adjacent street at the bottom of the garden.


    Either might invalidate your home insurance.

    >Always drive them in pairs, several feet apart. Fit bolted test links
    >in each nest so that you can measure the earth impedance (it can
    >change over time). Some amateur radio handbooks have tables of
    >number of rods needed for different impedances and soils. You don't
    >get much of an earth from one rod.


    The idea is that everything LOCALLY is protected by the same ground.
    By this I mean that the telephone, cable TV, wide-band network and
    everything else uses the same fault ground as the power. The idea is
    to keep the _difference_ in voltages within reason. This can be done
    by having all lightning protective devices grounded at the same ground,
    or by using a "whole house" surge protector, or merely the "whole
    computer" equivalent. Having the fault currents going to _different_
    grounds means that the grounds will be at different potentials, and
    that probably means those components you are trying to protect will
    be seeing excessive _differential_ voltages. That's bad news.

    Old guy

  12. Re: Scripting, UPS Selection, etc. (was: [O/T] Hans leads police to Nina's body)

    On Sat, 16 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , Adam wrote:

    >Moe Trin wrote:


    > myscript somefile
    >
    >does something with the file, and
    >
    > myscript -z somefile
    >
    >does the same thing and then something more,


    Minor comment - I tend to use something approximating meaningful words
    as parameters, rather than dash options. You _may_ run into quoting
    problems otherwise.

    >I imagine part of my script would look something like
    >
    > ZFLAG=initialvalue


    OK but the variable probably needs to be quoted if it's a string, not if
    it's a numeric assignment.

    > if [ $1 == '-z' ] ; then
    > ZFLAG=othervalue
    > fi


    OK, but that should be '[ "$1" = "-z" ]'. You can use an 'if else fi'
    construct.

    if [ "$1" = "-z" ] ; then
    ZFLAG="othervalue"
    else
    ZFLAG="initialvalue"
    fi

    If $1 is to be tested for more values, it is possible to use an 'if,
    elif, else, fi construct, but unless $1 is not a string, you are
    starting to reach in the point where a 'case' statement may be more
    efficient.

    > if [ $ZFLAG -op thirdvalue ] ; then
    > [do something more]
    > fi


    '-op' isn't a valid test, but other than quoting that's OK.

    >so what should I use for 'initialvalue', 'othervalue', 'thirdvalue' and
    >'-op', or what would most people expect to see there?


    You don't _have_ to set any - you could simply test "$1" for various
    values. As a common trait, using integers - 1, 2, 3, 4, etc. is common.

    >> That's mainly an experience thing. You _tend_ to use the commands
    >> you are familiar/comfortable with.

    >
    >True, though I seem to "discover" a new one often.


    Happens to everyone. ;-)

    >From your code snippet a few messages ago, I found out about 'mktemp'


    This is why I (and others) post code. We let people get new ideas. Some
    of them are definitely 'duh, is that all it takes'. ;-)

    >>> Either it's reading 5-8% low, or the bulb manufacturers are chiseling.

    >>
    >> Yes.

    >
    >Any idea which? I'd guess the meter reading is low, because all three
    >filaments (in two bulbs) measure roughly 95% of spec.


    It could be either. Lower wattage tends to increase the life, at a
    smaller cost of light output. Don't forget that the line voltage is
    allowed to be +/-5 percent, and if it's high, bulb life will be less.
    Long ago in another universe, we used to buy 130 Volt lamps because of
    the significantly increased life at 120 Volts.

    >[UPS sizes]
    >> The other concern *for me* is the amount of heat generated

    >
    >Not as big a concern for me, unless it's a lot of heat. 25 years ago I
    >had an external 5 MB HD that did a nice job of heating up the room.


    It shouldn't be that much. Yes, I know about those old drives.

    >Yes, because the cost of developing software is fixed, but the hardware
    >is per-unit. (Am I clear?)


    That's most of it. The other thing is if you are in a situation where
    features are changing over time - a software change is MUCH easier than
    hardware rework for the customer. Several years ago, GE was advertising
    that changing the rated power on one of their big jet engines was as
    simple as replacing a Cannon plug (unscrew, screw in the new one). That
    was actually changing software limits in the electronic fuel control.

    >> The 10:1 rule goes back quite a few years, and relates to consumer
    >> electric goods


    >Okay. My figure isn't scientific at all, just comparing "store cost"
    >and "retail price" on the shipping invoice.


    That may be valid, but it may also not include other "undocumented"
    cost reductions - the famous MSRP on the car verses the "dealer
    invoice" price - yeah, right.

    >That was 25 years ago; I don't know but I'd guess the markup is a
    >little less now, with all the competition.


    Things obviously have changed over the years, and comparisons may be
    more difficult. Some consumer electronics intentionally have different
    model numbers for different retailers, just to make comparisons more
    difficult. Marketeers.

    Old guy

  13. Re: Scripting, UPS Selection, etc.

    Moe Trin wrote:
    >> so what should I use for 'initialvalue', 'othervalue', 'thirdvalue' and
    >> '-op', or what would most people expect to see there?

    >
    > You don't _have_ to set any - you could simply test "$1" for various
    > values. As a common trait, using integers - 1, 2, 3, 4, etc. is common.


    I wasn't clear. In one script I'm working on, it takes one or more
    options specifying what actions to take, followed by some directory
    names to act on. ATM part of the script looks like:

    #!/bin/bash
    [snip]
    CATFLAG=1
    PARFLAG=1
    UNRARFLAG=1
    DELETEFLAG=1

    # parse command line
    PARSECMDLINEFLAG=0
    while [ $PARSECMDLINEFLAG -eq 0 ] ; do
    case "$1" in
    -c)
    CATFLAG=0
    ;;
    -p)
    PARFLAG=0
    ;;
    -u)
    UNRARFLAG=0
    ;;
    -d)
    DELETEFLAG=0
    ;;
    *)
    FINDDIRS=$*
    PARSECMDLINEFLAG=1
    ;;
    esac
    shift
    done

    [and farther down]

    if [ $CATFLAG -eq 0 ] ; then
    [do stuff]
    fi
    # note that there's no "else" here
    # if '-c' not specified, no alternate action taken

    if [ $PARFLAG -eq 0 ] ; then
    [do stuff]
    fi

    and similarly for UNRARFLAG and DELETEFLAG.

    As it stands, a flag value of 0 essentially means "yes" or "true", and
    non-zero means "no" or "false". Will I look like less of a novice if I
    use 0 for yes/true, or use nonzero for true and 0 for false? Or should
    I define TRUE and FALSE as integers, and then compare variables to $TRUE
    and $FALSE? Or is there some other way most people would handle this?

    I'd like to do whatever's considered common practice, so (a) I don't
    look like an idiot, and (b) it will be easier for others to understand,
    and therefore more likely they'll want to improve on it.

    [next topic: incandescent bulb wattage]

    >>>> Either it's reading 5-8% low, or the bulb manufacturers are chiseling.
    >>> Yes.

    >> Any idea which? I'd guess the meter reading is low, because all three
    >> filaments (in two bulbs) measure roughly 95% of spec.

    >
    > It could be either. Lower wattage tends to increase the life, at a
    > smaller cost of light output. Don't forget that the line voltage is
    > allowed to be +/-5 percent, and if it's high, bulb life will be less.


    I see that if a bulb is meant to be 60W @ 125V, then it
    would be about 260 ohms, and at my 121.6V would be 56.8W (95% of spec).
    My parents' house is about 116V, so the same bulb there would be about
    51.6W and not as bright, but would (on average) last longer.

    I know voltage also affects the color of the light from an incandescent
    bulb. Does anyone know whether the light output and color of
    fluorescent bulbs changes much when voltage varies within the range of
    110-125V?

    > Long ago in another universe, we used to buy 130 Volt lamps because of
    > the significantly increased life at 120 Volts.


    They used to make 130V lamps???

    >> Okay. My figure isn't scientific at all, just comparing "store cost"
    >> and "retail price" on the shipping invoice.

    >
    > That may be valid, but it may also not include other "undocumented"
    > cost reductions - the famous MSRP on the car verses the "dealer
    > invoice" price - yeah, right.


    My experience was at a chain where retail price was fixed by HQ.

    >> That was 25 years ago; I don't know but I'd guess the markup is a
    >> little less now, with all the competition.

    >
    > Things obviously have changed over the years, and comparisons may be
    > more difficult. Some consumer electronics intentionally have different
    > model numbers for different retailers, just to make comparisons more
    > difficult.


    I know; it makes comparisons almost impossible, especially on towers and
    laptops. And it's not just electronics. For a while I drove a Ford
    Taurus, and still don't know how it differed from a Mercury Sable except
    in prestige and sticker price.

    Adam

  14. Re: Scripting, UPS Selection, etc.

    Adam wrote:


    [snip]
    > I know voltage also affects the color of the light from an
    > incandescent
    > bulb. Does anyone know whether the light output and color of
    > fluorescent bulbs changes much when voltage varies within the range
    > of 110-125V?

    [snip]

    Yes. Somebody knows. :-)

    The energy (color) of a photon from a fluorescent light depends
    on difference between energy levels of electron orbits in the
    fluorescing material.

    No, the color does not change with voltage.

    As the voltage is reduced they tend to suddenly not work at all.


    --
    Peter D.
    Sig goes here...

  15. Re: Scripting, UPS Selection, etc.

    Adam wrote:
    > And it's not just electronics. For a while I drove a Ford
    > Taurus, and still don't know how it differed from a Mercury Sable except
    > in prestige and sticker price.


    The interior and some accoutrements were slightly different, and
    supposedly of higher quality. Advertising was oriented toward
    older more affluent customers. Platform, power train, and all
    basic components were exactly the same and from the same production
    lines.

    Cheers!

    jim b.

    --
    UNIX is not user unfriendly; it merely
    expects users to be computer-friendly.

  16. [O/T] Re: Scripting, UPS Selection, etc.

    Peter D. wrote:
    >> Does anyone know whether the light output and color of
    >> fluorescent bulbs changes much when voltage varies within the range
    >> of 110-125V?


    > The energy (color) of a photon from a fluorescent light depends
    > on difference between energy levels of electron orbits in the
    > fluorescing material.
    >
    > No, the color does not change with voltage.
    >
    > As the voltage is reduced they tend to suddenly not work at all.


    Thanks very much, Peter! Would you mind terribly much if I asked
    another question (probably an obvious one) about fluorescent bulbs, in
    particular the kind that screw into a socket meant for an incandescent
    bulb? If my incandescent lamp says, for example, "60W bulb max", is it
    safe to use a "more powerful" fluorescent, like one labelled "20W
    [fluorescent] replacement for 75W [incandescent] bulb"? (And why do the
    dollar-store fluorescents emit such visibly blue light? Is that why
    they're sold for $1?)

    Peter or anybody, feel free to reply on- or off-list, whichever you
    prefer. I realize this isn't exactly a Linux question, but it's not far
    from one -- only about one foot (30 cm) away on my desk. Thanks in advance!

    Adam
    --
    Email: adam seven zero seven AT verizon DOT net

  17. [O/T] Re: Scripting, UPS Selection, etc.

    Jim Beard wrote:
    >> And it's not just electronics. For a while I drove a Ford Taurus, and
    >> still don't know how it differed from a Mercury Sable except in
    >> prestige and sticker price.

    >
    > The interior and some accoutrements were slightly different, and
    > supposedly of higher quality. Advertising was oriented toward
    > older more affluent customers. Platform, power train, and all
    > basic components were exactly the same and from the same production
    > lines.


    So there isn't really much difference, mainly that a Mercury is
    higher-status than a Ford (but not as high as a Lincoln). Thanks, Jim!
    I'm continually impressed with the collective knowledge of this group.

    Adam

  18. Re: [O/T] Re: Scripting, UPS Selection, etc.

    On Mon, 18 Aug 2008 11:51:19 -0400, Adam wrote:

    > If my incandescent lamp says, for example, "60W bulb max", is it
    > safe to use a "more powerful" fluorescent, like one labelled "20W
    > [fluorescent] replacement for 75W [incandescent] bulb"?


    Your electrical ratings are what you pay attention to.

    You are not supposed to put/plug a device in, which draws more
    that what is specified.

    What they are telling you is, the 20W puts out as much light as a 75W bulb
    but only draws 20W of power. Whatever you put in the 60W fixture
    needs to draw less than 60W.

  19. Re: [O/T] Hans leads police to Nina's body

    In article ,
    ibuprofin@painkiller.example.tld (Moe Trin) in alt.os.linux.mandriva
    wrote:

    >On Fri, 15 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    >, David Powell wrote:
    >
    >>Jim Beard wrote:

    >
    >>>First, you need to ground your telephone line as well as the
    >>>electrical wiring.

    >>
    >>

    >
    >I'd tend to agree. Bolting either side of the phone line to ground
    >isn't the best idea. What _IS_ desired is to use devices - whether they
    >be spark gaps, MOVs, TranZorbs, or Surgistors - to clamp (limit) the
    >difference in the voltage between said lines and local references.
    >
    >>That may be possible in some countries, but it's bloody dangerous
    >>(and illegal) to do that in most European countries. Here, the phone
    >>line has a single earth point at the positive pole of the battery in
    >>the exchange.

    >
    >I'm not a telephone expert, but my understanding is that the lines are
    >not hard grounded. Certainly, tests on my home phones show the lines
    >are not well grounded.


    How can you tell? You can measure the earth impedance from your local
    earth to the phone line, but you cannot measure the impedance between
    phone lines and earth at the exchange, and, as you write later,
    there's the resistance of the lines to consider. Probably not that
    much difference between earths in areas with wet alluvial stuff under
    the top soil, quite different if it's granite underneath. Usually
    positive pole earthed, something to do with electrolytic corrosion.

    >>AC signals at the receiver are isolated by a transformer. All of the
    >>telephone wiring forms an equipotential zone for all subscribers on
    >>that exchange.

    >
    >I know what you are saying, but that doesn't appear to be the case here.
    >
    >>The power grid is earthed at the star point /neutral of the
    >>distribution transformer (usually 750-1000 kVA) and taken to the
    >>consumer in a five or three conductor cable. Any metallic service
    >>pipes are bonded to the earth conductor at every consumer's incoming
    >>terminals.

    >
    >Very different from the US. First, the supply transformer is local,
    >and because of I*R drop in the wires, is unlikely to be more than a
    >very few hundred feet - commonly less than 200 feet - from the
    >transformer to the customer. There is a _required_ ground at each
    >customer's demarc, and at the transformer. The ground is a safety
    >feature, and isn't meant to be carrying currents except during
    >fault conditions. Most residential distributions are a split
    >single phase (center tapped winding), providing 120 +/- 6 VAC
    >from each hot wire to the neutral. Between the (normal) two hot
    >wires, you get 240 +/- 12 VAC unless you are actually getting two
    >of three phases (208 +/- 10 VAC).
    >


    I accept all that, that's why I qualified my post to European
    practice. I'm aware of your practice with heavy current stuff, not
    domestic. Not much difference between a 13.8kV/~400V 3ph distribution
    substation and our 11kV/433V ones. However, that's not the point.
    The point is that neither of us has a single, nationwide, connected LV
    distribution system, it comes in finite chunks, and over the area
    served by telephone exchange copper there will be many such chunks,
    all with their own idea of what ground potential is.

    >>Earth the phone line at a subscriber's premises, guess where the
    >>current flows. If you're lucky it will just melt the incoming phone
    >>cable.

    >
    >Highly unlikely - the telephone is an energy limited device (line
    >impedances). The normal wires used depend on distances when they
    >were originally installed, but vary from #28 AWG (0.0126" dia, 64.9
    >Ohms/1000 feet) to #19 AWG (0.0359" dia, 8.05 Ohms/1000 feet).
    >


    It's energy limited wrt the 50V exchange battery. True only if the
    telephone system is used as designed, with no tampering by the
    subscriber. Introduce connexions to higher voltage systems and that
    protection is lost.

    >>Sometimes you can see this when a multipair screened phone cable, as
    >>thick as your wrist, is carried a few miles on poles. Somehow the
    >>screen gets earthed at both ends. Nothing really bad happens until an
    >>earth fault on the leccy transmission grid. Cable melts.

    >
    >Yes, that could happen - but we don't use screened cable - it's all
    >just twisted pairs. There may be a steel messenger wire molded into
    >the cable jacket, but these tend not to be grounded.
    >
    >>Some cables went out of a window in the main office block to
    >>temporary pre-fabricated offices alongside. What they didn't know was
    >>that the temporary buildings were powered from a transformer in the
    >>factory, a different transformer to the offices. We fell over this
    >>one some years earlier. Someone strung a 80 mA TTY cable across the
    >>works about half a mile to the KDF9s. This was done when the works
    >>were on their 14 day holiday shut down. (It worked well, no more
    >>mile walks in the rain to the terminal room.) The fire started at
    >>around 07:45 the day the works people returned, switched on their
    >>machinery, and caused a 20V difference between distribution
    >>transformer star points/earths.

    >
    >This indicates a power problem, as there should not be current flowing
    >in the ground leads. This also indicates a severe problem with idiots
    >allowed to run wiring in an unsafe manner.
    >


    Nope, sorry, you just don't understand the problem, nor power
    distribution. I have not suggested that currents flowing in the
    ground leads caused the fire, there's always some, it's usually
    harmless. Unless you have a perfectly balanced three phase system then
    there will be a neutral displacement and a neutral current, and a
    voltage drop on the neutral conductor. Between any two distribution
    systems fed by different transformers there will be a difference in
    neutral voltages. That is not the problem. The problem is that if
    you break the rules and connect the neutrals from two different
    transformers a current will flow in that connexion and there's no fuse
    in those circuits.

    What caused the fire was this. (apart from lack of knowledge of the
    distribution system) The connexion was mostly run in an armoured
    cable, mandatory at our factory if run in our cable trenches. At both
    ends it was terminated in a terminal box and the usual telephone type
    flexibles used with that sort of kit bundled in with many others and
    laced together. Most of the 20V appeared across the thin stuff and
    you can guess the rest.

    >>We had 12 different distribution transformers on the plot.

    >
    >So? I'm sorry, but I've got three three phase transformers in the
    >building I work in. There are telephone lines in the building coming
    >from one telephone room... in a different building. There are five
    >buildings on this site, each with multiple three phase distribution
    >points. We don't have grounding problems simply because the grounds
    >are not carrying currents. That's what the neutral line is for.
    >


    Just illustrating the problem. wherever you wanted to locate a tty on
    that plot then over 90% of such locations will be on different
    transformers.

    >>Never earth the screen of a serial link at both ends.

    >
    >Agreed
    >
    >>It can happen to you at home, transformer service areas do have
    >>boundaries. Probably OK to connect to the guy next door, but not to
    >>the guy on an adjacent street at the bottom of the garden.

    >
    >Either might invalidate your home insurance.
    >
    >>Always drive them in pairs, several feet apart. Fit bolted test links
    >>in each nest so that you can measure the earth impedance (it can
    >>change over time). Some amateur radio handbooks have tables of
    >>number of rods needed for different impedances and soils. You don't
    >>get much of an earth from one rod.

    >
    >The idea is that everything LOCALLY is protected by the same ground.
    >By this I mean that the telephone, cable TV, wide-band network and
    >everything else uses the same fault ground as the power. The idea is
    >to keep the _difference_ in voltages within reason


    Simply not possible. If your residential transformer has a service
    radius of 200 feet, and your telephone exchange 10000ft, there will be
    many transformers, each with its own local earth, for each exchange.
    (let's forget about subscribers who pay for separate feeds from two or
    more exchanges, for security) I'm pretty sure that you don't run
    copper strip between all those transformers to tie the grounds
    together. We solve that problem by having separate earths and
    equipotential areas for the two services. It doesn't matter if the
    telephone lines are displaced from power ground, so long as there's no
    low Z connexion between. However, as I can see that you understand
    that the grounds can be at different potentials, it's difficult for me
    to see why you cannot accept that a destructive current can flow in a
    connexion between the two.

    >. This can be done
    >by having all lightning protective devices grounded at the same ground,
    >or by using a "whole house" surge protector, or merely the "whole
    >computer" equivalent.


    Not sure about that, usually best to take the lightning conductors to
    ground by the most direct route. Lightning doesn't go round corners
    very well.

    > Having the fault currents going to _different_
    >grounds means that the grounds will be at different potentials, and
    >that probably means those components you are trying to protect will
    >be seeing excessive _differential_ voltages. That's bad news.
    >


    You gave a description of your residential domestic supplies, I'm
    guessing the transformer is pole mounted. Much the same as our rural
    stuff, but ours is 3ph&n. We too have two earth systems. The earth at
    the pole is there to earth the transformer tank, which may take fault
    current and voltage at HV from a short inside the tank. The one at
    the consumer's terminals is the neutral to earth bond on the LV
    circuit. We need separate earths, the one by the pole may transiently
    rise to a voltage much higher than you would want in the house.

    Regards,

    David P.


  20. Re: Scripting, UPS Selection, etc.

    On Sun, 17 Aug 2008, in the Usenet newsgroup alt.os.linux.mandriva, in article
    , Adam wrote:

    Moe Trin wrote:

    >> You don't _have_ to set any - you could simply test "$1" for various
    >> values. As a common trait, using integers - 1, 2, 3, 4, etc. is common.


    >I wasn't clear. In one script I'm working on, it takes one or more
    >options specifying what actions to take, followed by some directory
    >names to act on. ATM part of the script looks like:


    ># parse command line
    >PARSECMDLINEFLAG=0
    >while [ $PARSECMDLINEFLAG -eq 0 ] ; do
    > case "$1" in


    > *)
    > FINDDIRS=$*
    > PARSECMDLINEFLAG=1
    > ;;


    OK, so the directory list is mandatory, and is the last item[s].

    > esac
    > shift
    >done


    OK - this _could_ get you in trouble, if you fumble-finger an option
    other than "-[cdpu]" - that would be taken as the directories to parse
    and bails from the case statement. Off the top of the head, I'd
    rework the wild-card as a sanity check, and do the last bit outside of
    the case statement, but before the shift - perhaps

    -*)
    echo "Unrecognized option $1"
    exit 2
    ;;
    esac
    if [ `echo $1 | cut -c1` = "/" ] ; then
    FINDDIRS=$*
    PARSECMDLINEFLAG=1
    elif [ `echo $1 | cut -c1-2` = "./" ] ; then
    FINDDIRS=$*
    PARSECMDLINEFLAG=1
    else
    echo "Usage: $0 -[cdpu] directory"
    exit 3
    fi
    shift

    That's untested, but I suspect you see the idea. Note that I'm testing
    for an absolute _and_ relative path to the directory, and assuming that
    the user did make the last argument[s] to be the directory[s], though
    the two sanity checks mitigate that.

    >As it stands, a flag value of 0 essentially means "yes" or "true", and
    >non-zero means "no" or "false". Will I look like less of a novice if I
    >use 0 for yes/true, or use nonzero for true and 0 for false? Or should
    >I define TRUE and FALSE as integers, and then compare variables to $TRUE
    >and $FALSE? Or is there some other way most people would handle this?


    No, this looks fine.

    >I'd like to do whatever's considered common practice, so (a) I don't
    >look like an idiot, and (b) it will be easier for others to understand,
    >and therefore more likely they'll want to improve on it.


    This is fine, and you're not even eligible for the UUOC award ;-)

    >> It could be either. Lower wattage tends to increase the life, at a
    >> smaller cost of light output. Don't forget that the line voltage is
    >> allowed to be +/-5 percent, and if it's high, bulb life will be less.

    >
    >




    >I see that if a bulb is meant to be 60W @ 125V, then it would be about
    >260 ohms, and at my 121.6V would be 56.8W (95% of spec). My parents'
    >house is about 116V, so the same bulb there would be about 51.6W and
    >not as bright, but would (on average) last longer.


    That would be the "hot" resistance (which you have to determine by
    math, as an Ohmmeter won't provide the energy to warm up the filament),
    but this isn't unreasonable.

    >I know voltage also affects the color of the light from an incandescent
    >bulb. Does anyone know whether the light output and color of
    >fluorescent bulbs changes much when voltage varies within the range of
    >110-125V?


    The color of a fluorescent lamp is a function of the phosphor, or the
    gas _if_ it's a discharge type of device.

    >> Long ago in another universe, we used to buy 130 Volt lamps because
    >> of the significantly increased life at 120 Volts.

    >
    >They used to make 130V lamps???


    GE and Sylvania as I recall.

    >I know; it makes comparisons almost impossible, especially on towers
    >and laptops. And it's not just electronics. For a while I drove a
    >Ford Taurus, and still don't know how it differed from a Mercury Sable
    >except in prestige and sticker price.


    Jim Beard has answered that one - GM was notorious for this, but we're
    also ignoring cars made in other countries that have domestic labels.
    Izuzu, Suburu, Mitsubishi, and Mazda in Japan, Opel in Germany and
    others were sold as US labeled vehicles (for example, the Dodge Colt
    and Ram D50 were a Mitsubishi vehicle, as was the Plymouth Arrow).

    Old guy.

+ Reply to Thread
Page 6 of 14 FirstFirst ... 4 5 6 7 8 ... LastLast