A couple of weeks ago I was asked what kind of scanner to get to be able to listen to police and fire in Keller.  My answer was that it’s a bit complicated at the moment, and likely to be expensive, depending on how things shake out in the future.  This post is an attempt to add some detail to that and to provide this information to anyone else who may be interested.

Background and Basics
The first thing that you need to know is that Keller is a participant in the Northeast Tarrant County Motorola 800MHz trunked system. 

Trunked?  What’s the heck is that?  It can be a bit complicated, but here is the basic explanation.

Trunking is a way of using a set of frequencies in a more efficient manner than the old conventional way. In the past, each agency in a political sub-division would get some frequencies licensed to them and then would use those discrete frequencies just for themselves. Since there is a finite amount of available radio spectrum, as the number of users increased crowding and interference occurred.

…

Trunking is the use of several repeaters, on different frequencies in the same band, operating together under computer control to allow the pooling of resources for several agencies.

The trunking radio, in a patrol car for example, is much more sophisticated than the simple transceiver previously used in a simplex or repeater configuration. The trunking transceiver is a frequency agile radio capable of receiving instructions and changing frequencies on the fly. All trunked radios operate in a similar manner although the type of trunking technology used by each type of trunked radio system differs greatly.

In the trunked radio environment, each agency is assigned one or more talkgroups that the agency’s communications will use. All agencies on the system will have different talkgroups but all will share the same pool of frequencies.

I’d suggest reading the entire article I linked above from the Radio Reference wiki, as it’s important to understand the concepts if you’re going to attempt to program a scanner (your scanner’s manual will reference some of the same vocabulary, such as “talk groups” and “control channels”).

The Nextel Interference Problem/Rebanding
If it weren’t for Nextel, you’d be able to go down to Radio Shack or some of the online dealers and buy yourself a “trunk tracker” scanner.  After a little programming (which some of the dealers will do for you), you would be able to listen to everything.  And, in fact, you could still do this today.  The problem will occur in the near future, when your trunk tracker will stop working if you don’t understand what’s changing and how to accommodate the changes.

So what does Nextel have to do with this?  It turns out that those silly little walkie-talkies aren’t really a cellular service, but are instead classified by the FCC as an “Extended (or Enhanced) Specialized Mobile Radio” service (ESMR).  They are the largest ESMR provider, and are licensed in the same frequency ranges as public safety repeaters (in the 800 MHz band).  The interference problem occurs because their ESMR transmitters may be pushing up to 3kW, while public safety systems are operating at between 125 and 300 Watts.  Even though they are on different frequencies, they’re close enough that the much higher power of the Nextel signals can completely obliterate the lower-power police and fire transmissions.

To fix all this, the FCC has redesignated the spectrum allocation so that ESMR has a contiguous block far enough away from the public safety frequencies so as to prevent interference.  Nextel will be paying for the rebanding and in return is being given some RF spectrum in the 1.8 GHz range.

What this means for scanners
In order to follow a trunked system, scanners must understand the type of system and be able to decode the control channel to determine which frequency will be used for a transmission on a given talkgroup.  In some trunked systems, rather than send the actual frequency in the control channel data they use an index into a frequency table (this reduces the amount of data that has to be sent over the control channel).  When rebanding occurs, the frequency tables will change.  Some scanners have these tables hard-coded in ROM and will not be able to be changed.  These scanners will quit working with trunked systems when those systems have been rebanded.  Others have the tables in flash-upgradeable firmware.  Provided that the scanner’s manufacturer provides a firmware upgrade, these scanners should be able to work after rebanding. 

Unfortunately, many of the most affordable scanners fall into the non-upgradeable category.  So finding a scanner that will work today and into the future has become more expensive (due to having to get the newest and/or more high-end models).  While Nextel is reimbursing the public safety licensees, they have no obligation to pay us for new scanners. 

How does this affect Keller?
The upgrade will happen in phases.  The first phase, to be undertaken in the very near future (within the month I’m led to believe) is to replace all of the handhelds and all of the mobile units with newer models that can handle the new trunking scheme.  However, the radios will still operate on the existing system with the existing frequencies.  The second phase (which I don’t have any dates for) will be to actually switch frequencies.  This will be the watershed date for local scanner enthusiasts. 

Another spanner in the works: digital
But to complicate things, Keller’s new radios will all be digital capable.  While I have not heard any definite plans for switching to digital voice, the possibility of it means that a scanner purchaser should think long and hard about buying a digital trunking scanner.  The problem is that digital trunking scanners command a serious premium over their analog-only cousins at the moment (I will discuss models and prices below).

For those that do not know about these things, the difference between conventional (analog) and digital trunking is in how the voice signal is sent over the air.  In all trunking systems the control channel is digital (think of it as sort of like a modem signal) because it’s sending information about channels.  In conventional systems, though, the actual voice data is analog (i.e. generally narrow-band FM).  In digital trunking systems, the voice signal is converted to a stream of bits which is then modulated onto the voice channel (think modem again).  The advantage to digital is that additional checksum data can be sent to allow for recovery of small errors in the stream due to noisy signals.  Which means that digital voice can be much clearer than analog (it completely takes out static, pops, hiss, etc).  The disadvantage is that digital can only compensate for a small amount of lossage before the whole signal is lost (at which time you get nothing on the receiving end as compared to analog where the human ear may still be able to pull some information out of the noise).

It should be noted that digital does not mean encrypted.  It’s certainly a component of modern encrypted systems, but the majority of public safety systems that are using digital are not encrypted. 

So what scanner should I get?
That’s harder to say, as you will have to make the tradeoff between cost, ability to upgrade, and digital capability.  You will also have to consider which form-factor you want: desktop/mobile or handheld. 

There are only two major brands of scanners on the market in the United States: Radio Shack and Uniden (Bearcat).  Most (not all) of Radio Shack’s scanners are made by a company called GRE, which recently started marketing their scanners directly.  The Radio Reference wiki maintains lists of both Radio Shack and Uniden scanner models, their capabilities, and whether they can be upgraded for rebanding.  GRE is new enough on the U.S. market that all of their trunking scanners are advertised as being rebanding capable. 

Most scanners will also list in their description how many trunked systems they can follow at once.  For example, you will see phrases like “triple trunking” or “digital triple trunking”.  This means that it can follow up to three trunked systems at the same time (for example, my old Radio Shack scanner is programmed to be able to receive both the NE Tarrant system as well as the Ft. Worth system, although I have to turn Ft. Worth *off* to have any chance at all of hearing what’s happening in Keller because Ft. Worth has hundreds of talkgroups and thousands of users who seem to be talking 95% of the time). 

There are lots of online dealers out there, but I’ve had good luck so far with Scanner Master.  They carry all lines of scanners (Radio Shack, Uniden, and GRE), accessories, and (most importantly for the newbie) they offer programming services.  Even if you get the programming service, I’d recommend getting the software for the scanner and all the cables to hook up to a computer, as you can download info from Radio Reference to set up the scanner the way you want it.  As noted above I have multiple systems set up, but I also have variations on those system stored so I can limit things to a manageable level.  I have a config for the “full” NE Tarrant system (all talkgroups), one for NE Tarrant with just the Keller police and fire talkgroups, and another for Ft. Worth.  I can switch between them with the press of a button, depending on where I am and what I want to hear.

Be warned, though, that a digital trunk tracking scanner will start around $499.  If you’re willing to live without digital, trunk tracking scanners can be found in the $200-$250 range (but be sure to check that the model you choose will be able to be reprogrammed for rebanding).

Finally, the legal stuff…
Some people have the idea that scanning is illegal or some sort of violation of privacy.  Others may object to my giving the above information as it may help criminals.  First, at the Federal level the law is such that as long as you are not attempting to receive cellular or other telephone communications (i.e. cordless phones), you can generally listen to any non-encrypted transmission that you can receive.  All modern scanners come with the cellular frequencies blocked.  If a transmission is encrypted, then you may *not* attempt to decrypt it.  Each state has its own laws regarding scanner usage (both stationary and mobile), so you will need to check for yourself to see what you may or may not do.  A good resource for the laws in each state is the Scanner Laws site.  I am not a lawyer, though, so you should confirm all legal issues yourself before proceeding.

As for assisting criminals, all of the above information is freely available to the public, so I am not publishing anything new or providing anyone with special insight, other than distilling certain information into one place.

This situation is why I don’t like to let anyone auto-renew or auto-bill to my bank account.  In this instance I’m using a credit card, which offers more protection.

We are currently aware of some billing inaccuracies which we will be resolving later today. All of these issues will be resolved, and any billing errors corrected. We sincerely apologize for any hassle this has created.

– UPDATE Tue Jan 15 04:13:17 PST 2008 –

The problem has been found and corrected. We are currently going through all erroneously charged accounts and refunding money. Stay tuned for further updates.

– UPDATE Tue Jan 15 07:25:16 PST 2008 –

Scripts are still running to correct the original problem. Last night, this guy ran a standard billing cycle to clean up stragglers from 2007. Unfortunately, the biller was run for 2008 (December 31st, 2008 to be exact). This caused everyone to be billed as if today was 2008-12-31, wreaking the havoc that we are so sorry you had to be put through.
…

The reason I noticed was that I got an “automatic rebill” notice this morning for my hosting account, which was weird, since the account normally renews in April.  But what really caught my eye was that it billed me TWICE:


We have the last payment on this account to be $1.99 on 2008-01-15 04:40:27.  (Ed: I think this is a dividend payment, as I didn’t pay them anything myself.)
Since then the following charges have been made to the account, which you have just paid for:

      2008-05-01 - $287.40 for “Code Warrior” through 2009-04-30.
             
      2008-05-01 - $287.40 for “Code Warrior” through 2009-04-30.
             

For full information on your previous balance please visit:
< Panel Invoice subtab >

SUMMARY OF PAYMENT

Amount: $534.54
Credit card: *XXXX (ex XX/YY)
Payment code: #######

 

Charges on your credit card bill will be from “DreamHost.com”.


It was only after I tried to contact billing that I realized that there must have been a site-wide screwup.  The email got bounced as undeliverable due to “too many connections.”  Which is understandable, considering that a lot of people were using debit cards linked to their checking accounts, and a lot of these people woke up to overdraft fees and bounced checks this morning because of the unexpected charges; others had their hosting accounts suspended for “past due” balances if they didn’t have any automatic form of payment on file.  As of this moment, there are 576 angry comments on the above post, and the number keeps growing. 

This comment shows some of the pain that can come from unexpected charges of this sort:

As a (until now) happy customer of dreamhost this mistake is extreme anger inducing. I have been charged for a full year right before my mortgage payment for the month cleared… making my mortgage payment check bounce.

lets see:
$119.40 yearly hosting fee
$50 bounced check fee
$100 missed mortgage payment fee
30 day late on my credit record (which kills my refinancing hopes for this year)

WAY TO GO DREAMHOST!

Ouch!

While I’ve been fairly indulgent of Dreamhost’s failures over the years because they offer inexpensive hosting with lots of features, this sort of billing screwup isn’t something that I can so easily overlook.  It either reflects a lack of appropriate financial controls, or an overly cavalier attitude towards the billing process.  Either way, I am strongly considering removing my credit card information from their system and doing all renewals manually. 

If it’s this easy to screw something up by accident, imagine what someone with ill-intent could accomplish.  So I would suggest to Dreamhost that they not only need to fix the problem, but they will need to go above and beyond a simple fix to reassure customers and to regain lost trust.  That would include:

• Compensation for any and all fees or costs incurred due to the erroneous billing (i.e. returned check fees, overdraft fees, etc)
• A public audit of their billing processes, with a published report to all customers of the outcome.
• A published plan that lists, item-for-item, fixes for all issues identified in the audit.

Updated after first publish to add quote from comment thread on DreamHost Status.

I don’t know what percentage of people get their TV programming OTA (Over The Air; i.e. direct via antenna), rather than via cable or satellite.  But I suspect that in terms of absolute numbers the number of TV’s out there that are hooked up to rabbit-ears or external antennas is fairly sizable. 

While most of my viewing is done on one of the TVs that are connected via FiOS, I also have a small one mounted on the wall in the kitchen that I use to watch the morning news (and sometimes the noon news).  I never bothered to have it hooked up to cable, but instead just use rabbit ears.  The only stations that come in very well are the local Fox and ABC channels (4 and 8), but that’s generally sufficient for my news fix.

Unfortunately, this TV is not going to be able to receive any OTA broadcasts after 2/17/09 because of the transition from analog to digital.  The options appear to be to get a digital converter box or to get a new TV with an ATSC tuner.  There is a government program to offer each household up to two $40-off coupons good towards the purchase of converter boxes.  However, there are not many converter boxes available yet, and ones eligible for the program won’t be available until February or early March, so I don’t know whether this coupon will be worth the trouble or not.  Heck, the TV itself only cost me $59, so it seems silly to spend very much on a converter for it.

On the other hand, searching for small TVs with ATSC tuners turns out to be an expensive proposition.  All of the small digital TVs appear to be flat-screen (generally LCD), and these tend to start around $200, which is more than I want to spend for a TV in the kitchen. 

There is one other alternative I’m considering: a wireless audio/video remote “extender”, connected to the FiOS box in the living room.  This would allow me to view all of the cable channels on the existing TV as long as I didn’t try to tune to any HD channels.  In addition to skipping the converter box, it would also improve the picture quality over the rabbit ears (or it should).  There are even units that include a remote-control IR channel (the receiver at the TV end includes an IR receiver/transmitter that sends the signal back to the “sender” at the converter box end, so you can control the remote converter box from the local TV).  The big problem, though, is that most of these are 2.4GHz, so it will be difficult to find one that won’t step on my wireless network.

Hmm…  here’s a unit that operates at 5.8GHz.  Unfortunately, it’s $250.  For that kind of money I could get an LCD TV.

Anyhow, this also brings up some interesting political/philosophical issues.  From a libertarian standpoint, the government shouldn’t be handing out “our” money for these sorts of things.  Actually, if you follow the logic, the government shouldn’t be mandating the digital transition.  If it was a good idea, and profitable, the market would sort it out. 

Also from a libertarian standpoint, if you realize that the program exists and there isn’t anything you can do to stop it, do you then try to take advantage of it?  Do you rationalize it as trying to get a small refund on the thousands of dollars of taxes you’ve already paid?  Or is the principled stand to refuse to participate?

Being a gadget junkie, I love hearing about these kinds of things.

Just 12 months after demonstrating a working prototype of the world’s smallest projector, Redmond, Wash.-based Microvision is unveiling a full functioning, self-contained prototype that should be available as a real product—possibly from Motorola—later this year.

Dubbed SHOW, the lensless PicoP projector is designed for the home and business use, and uses tiny lasers to shoot a WVGA (848 by 480, roughly DVD resolution) image on virtually any surface that isn’t a dark color or textured. It can even project onto curved and uneven surfaces. So, from a distance of two feet, it could project a two foot diagonal, full-color image on a white T-shirt. From five feet away, it could show a five-foot image on, say, a white wall or ceiling.

This thing isn’t much bigger than a deck of cards:

It would be great for people who travel for business, since it means not having to lug a projector around (even the travel ones are quite a bit bigger than this).  But beyond that, I could see some additional uses.  Take an ultraportable PC (an oQo or perhaps even something like the Nokia 880) and hook one of these up along with a fold-up bluetooth keyboard, and you could have a portable desktop-like system without all the bulk (you could probably carry it in your pocket; if you had a fairly large pocket).  Or, since the projection engine is so small, it could even be embedded in laptops or ultraportables to begin with. 

It accepts input from any video-capable device, so (as shown in the picture), you could take your movies or pictures along on an iPod and show them without having to use a TV or carry a large projector.  The projected price is given in the article as $200 to $300 (which is a steal compared to the prices of current projectors; although to be fair the other projectors have better resolution).

Linked via Slashdot.

Delivering good software can be a challenging process, especially so when numerous development and design principles have been ignored or violated.  I was recently reminded of this due to my exposure to Verzon’s new FIOS TV Interactive Media Guide.  A month or so ago I received a letter telling me that I would soon be seeing a new interface on my DVR.  The day before the roll out I got a phone call telling me about the coming “upgrade” and pointing me to the website for a tutorial.  This should have been a warning to me that something was up. 

I’ve been in the software business for about 14 years now, doing everything from tech support to programming to systems design. Over that time I’ve learned a thing or two about delivering software as well as about UI (User Interface) design.  While UI is not my specialty, it’s something that I have to take into account while doing design work.  Whenever you have a system that is going to be used by people (rather than other systems), then you have to think in terms of the user while doing your design.

Anyhow, I told you that so I could tell you this:  It is my considered opinion that the Verizon FIOS TV Interactive Media Guide as it exists today is an unmitigated piece of crap.  The UI is so flashy as to be difficult to use and distracting.  It’s slow. It’s unreliable.  It’s inconsistent.  It represents a failure in requirements analysis, ignorance of UI design principles, and its implementation and roll out demonstrates an utter lack of understanding or consideration of UCD principles. 

Lest you think I’m just bloviating, allow me to give a few examples.

The first (and most egregiously annoying) mis-feature is the “fixed focus” program guide.  Here’s what it looks like:

As you press Up or Down Arrow the “highlight” stays fixed and the guide data moves underneath it.  It’s kind of like looking at a specimen through a microscope by moving the slide around underneath it.  Unfortunately, I find this highly distracting and somewhat nauseating.  First, I tend to think spatially.  This means that I think of my local channels as being “up there” on the guide, things like Discovery (channel 100) being “in the middle,” and the HD channels being “down there” (800’s).  Scrolling through with the old program guide was kind of like navigating in a video game, in that I could correlate spatial position to channels.  My brain is also accustomed to scanning the entire page at once.  My usual Sunday morning activity is to open up the guide, enter 100, and then scan across the Discovery channels to see if there’s anything interesting on.  And I do this a page at a time.  It’s kind of like a visual “grep.”  With the new guide, the huge visual wart in the middle interferes and causes me to have to refocus on the top and bottom halves separately, which interrupts my visual comparator and makes me irritable. 

Anticipating objections, though, I will admit that this is a personal thing.  There might be people who like looking at things through a microscope.  I’m just not one of them.  Still, I bring it up because there is no option to use the old style of guide.  This indicates to me that they didn’t talk to any users during design, didn’t talk to enough users if they did talk to any, and/or ignored feedback from the initial victimstesters.  A good design would have take human factors into account and would have either not used the fixed focus method or would have allowed for an alternate guide format.  More on this towards the end…

They have implemented a “tabbed navigation” system for the menu system, which is shown above.  This system has several issues, mainly in being inconsistent and violating the principle of least astonishment.  For example, pressing Right Arrow causes the next level menu to be activated.  The interface shows you the options in the next level menu as it’s displaying the upper level.  Unfortunately, if the list is long it attempts to center the list and the arrow points to something in the middle.  When you press Right Arrow to go into the next level menu it takes you to the top of the list, rather than the item to which the arrow was pointing.  This is surprising and annoying.  It should always go to the item that the arrow is highlighting. 

Further, it’s not consistent, as when you get to “leaf” items (i.e. lowest level items in the menu), pressing Right Arrow does nothing.  You have to highlight the item and hit OK.  The old version of the IMG let you use Right Arrow, so not only is the new version inconsistent, it removed previously useful functionality. 

The interface is also horribly slow and it provides no feedback that a request has been received or that an operation is under way.  This is a cardinal user interface sin.  Both the guide and the ‘My DVR’ function take 1.5 to 2 seconds after the button press to appear on my system.  There is no change in cursor, no change in interface color,  nor is there a sound to indicate that the button press has been received and that the data is being gathered.  I forget where I originally saw this, but most UI guides require acknowledging a button press within 0.1 seconds.  The operation can take longer, but you must give an indication that you received the button press and that the operation is under way.  Of course, the first thing that comes to mind is to use some sort of sound, but that gets annoying quickly (it was the first thing I turned off back when I had a Tivo).  So there has to be some sort of visual indicator on the screen to show that a button press was received and that an action is under way. 

This can be more than just an annoyance, by the way.  Users, thinking that nothing has happened, may be tempted to press the button a second time.  If the particular button can cause a selection to be made and then cause some action on that selection, there may be trouble.  The old version of the IMG had this problem, in that it (like its successor) did NOT acknowledge button presses and would often appear to be hung.  I accidentally deleted a couple of recordings by hitting OK too many times, thinking that the system had missed the first button press.

There are examples like this throughout the IMG, and it would take me forever to enumerate them.  Fortunately, someone else has already done so on the FIOS TV forum of DSLReports: IMG Bugs and Missing Features FAQ

I mentioned earlier that there didn’t appear to be much user feedback incorporated into the design.  I find this to be the most egregious sin of the entire project, and it’s not so much a failure of the development team as it is of the Verizon management.  I say this because they have been rolling this thing out for months and they have known since the first deployment that it was bug-ridden and that many people hated the new guide.  Rather than step back to incorporate the feedback, they continued rolling out the system.

It’s clear to me that either Verizon management wasn’t hearing the same story as their customers were telling, or they simply didn’t care.  On August 3rd, John ‘CZ’ Czwartacki (Executive Director – External Communications) posted an article on the Verizon PolicyBlog about how they were generating such great buzz over the new FIOS TV IMG with bloggers.  This lasted until the 9th, when frustrated customers found the blog entry and deluged him with over 100 comments about the ways in which the new IMG sucks. 

But it was this response that told me that Verizon management doesn’t get it:

UPDATE (9/18/07): We hear you!  Thanks for all the great comments and suggestions.  We’ve incorporated many of them into a maintenance release that we intend to deploy shortly after we finish deploying version 1.0 across the country.  Keep the feedback coming.

So they intend to give everyone the crappy IMG before they send out the fixes.  It seems pretty stupid to me to expose your customers to a bad, buggy, interface and then fix it (especially if the new IMG can delete recorded programs, mess up scheduled recordings, cause random reboots, and generally wreak havoc with your settings).  It sounds like someone let the programmers make that decision, rather than doing the right thing and making a new combined image with all the fixes rolled in.  Given how long this thing has been out there, and how loud the screaming has been, I should never have received such a bug-laden piece of crap on my DVR at the end of October.  That’s simply inexcusable.

I’ve been wanting to buy high-quality non-DRM’ed digital music in MP3 (or some other fairly open format) for quite a long time.  Searching my archives I see that I first brought this up four years ago.  And after using iTunes music store for a while, I finally got tired of the DRM about three years ago.  Recently, Apple announced DRM-free music, for a slightly higher price, although it was of slightly higher quality.  I played with it a little bit, but didn’t do too much with it, as it was still in Apple’s format (although being DRM-free it would at least play on my Squeezeboxes ) .

So upon learning that Amazon.com was offering DRM-free MP3 downloads in 256kbps VBR I decided I had to check it out.  First, installing their download app was pretty painless (although I have not used it to buy a full album yet).  The prices seem reasonable ($0.89/track in most cases).  And the selection is far better than any of the other DRM-free services I’ve investigated in the past. 

In fact, I fear that this could become quite a dangerous feature of Amazon.com.  It’s easy to lose track of what you’re spending when it’s only $0.89 a track.  I’ve already picked up one track that I couldn’t previously find without either buying the whole CD (or buying a DRM-encumbered version in iTunes).  That’s exactly the scenario that I think this setup is perfect for.  I would not have bought the whole CD for the one song, so ultimately the label and artist get some small amount of revenue that they would not previously have gotten, and I get a copy of the song to keep.  Classic win-win.

The best find so far?  Carl Douglas – Kung Fu Fighting (disco flutes and all).  No, I didn’t buy it, though.  I was just amused to find it ranked as the 16th most popular MP3 download in the “Dance & DJ” category.

It’s bad enough when you call a company and end up having to speak to a damn machine.  But having a machine call me and demand that I speak to it is beyond the pale.  I picked up a call late last week and was immediately greeted by a computer-generated voice that said, “Is this Aubrey Turner or spouse?  Please say ‘Yes’ or ‘No’.”  My immediate response was, “Hell, NO!  I’m not talking to a damn machine!” after which I hung up.  I looked up the number later and it turned out to be from my insurance company’s “informed health line” (basically a library of health information).

When I got home last night there was a message on my machine from the insurance company’s computer again.  Since I didn’t have the opportunity hang up it got around to mentioning what it wanted, which was for me to participate in a survey. 

So, here’s a note to Aetna:  I don’t participate in surveys, so don’t bother calling again.  Regardless, though, if you have need to call me, please have the common decency to have a human on the line.  I am not going to talk to your damn machine.

Sometime early next year the state plans to fully deploy a system for checking the validity of insurance cards.  The data will be available to police when they run a license-plate check.

Under the system, insurance information will pop up on the in-car computers of police officers when they check license-plate numbers. The state will pay for the $3.5 million-per-year system through a $1 fee on vehicle registrations. Motorists will still be required to carry their insurance card.

The first thing that came to my mind is that some poor schmucks are going to get hurt badly by this if they happen to be unlucky enough to be the victim of poor data in the system, especially in those jurisdictions where they impound your vehicle on the spot for failing to have insurance.  It appears I’m not alone, because it’s happened already in other states.

“The system has many flaws in other states where [it] has been enacted. Motorists with valid insurance have been pulled over, fined, or sometime had their cars impounded because the [correct] data does not match what the state has” in the database.

—Sandra Helin, spokeswoman for the Southwestern Insurance Information Service, an industry group

The state claims that there will be a “data clean-up phase” conducted before the system will go online.  But it doesn’t inspire confidence that the system provides no way for you to verify your own data.  You just have to hope that your insurance company correctly spots any inconsistencies before it goes online.

What peeves me the most about these sorts of systems is that there is never any accountability for screw-ups.  The onus is put on the victim (guilty until proven innocent) to prove that the system was wrong.  And in some cases, they just don’t care.  I’m reminded of the case where a woman was issued two (duplicate) tickets by the City of Garland from a red-light camera because she ran a red light.  The only problem with the citation was that she ran the light because an officer was blocking it to let her pass.  She was driving in a funeral procession.  In fact, it was her own mother’s funeral.  It took multiple attempts to get in touch with anyone at the city who could help, and when they did find someone they demanded a copy of the obituary.  But once the obituary was sent, she couldn’t get any information out of the city as to whether the tickets were really dismissed.  Worried that perhaps these tickets might cause warrants to be issued for her, she finally called Saul Garza at Fox 4 News.  Interestingly enough, the city managed to dig up someone who was helpful once a reporter started digging.

I’d like to see some real penalties for the people running the system if they cause an innocent person to be fined or have their vehicle impounded.  A system like this, if it can result in any sort of penalty against a person must be as close to 100% accurate as possible.  There also have to be safeguards and checks-and-balances so that people aren’t bounced between the vendor and the state when they try to get answers.  The red light cameras have really brought these issues to light recently, with people being fined for crossing an imaginary stop line or for vehicles they don’t own.  Unfortunately, in most cases, when attempting to correct the problem the victim gets bounced between the camera vendor/operator and the city and no one will step up and own the problem.  Perhaps if the city, state, and/or vendor in these cases would have to pay a hefty fine for each uncorrected mistake, they’d be more inclined to take responsibility instead of pointing fingers.  It shouldn’t require calling an investigative reporter to get problems fixed.

I tend to leave my PCs running for convenience.  Even with an XP desktop system I’ve been known to go two months without a reboot.  But the latest version of Firefox is really testing my ability to do so without having to carefully manage my applications.  Consider this:

That’s with six open windows running since Thursday (approximately 5 days).  There’s absolutely no good reason for a web browser to be sucking up nearly 1GB of RAM.  Up until I switched to Firefox 2.0 I didn’t have this problem.

I’ve done some searching in Google and there are some pages that suggest lowering the overall page cache, but there are others that say it won’t make a difference.  I really started noticing this on my old laptop as it would get horribly sluggish after a couple of days.  I’d check the running processes to find that Firefox was using 350MB or more, which is really bad on a system with 512MB.  I only noticed it on this system today because opening new windows was becoming very slow (this system has 2GB, which seems to have masked the problem somewhat).

I’m going to give the memory cache trick a try.  Hopefully it’ll do the trick.  Otherwise I either have to switch back to IE (I hate IE 7’s interface) or restart Firefox every other day.

Update:  Looking at some of the threads about memory leaks with Firefox I saw that the first response was always to have you delete all extensions and then see if the problem still occurred.  If not, then add the extensions back one by one until you found the culprit.  In this case, it turns out to be v1.5.0.32 of the PayPal Virtual Debit Card beta.  They only recently added support for Firefox, and I suspect that they did something naughty in their code that’s causing a memory leak.  I’ve submitted a report so that hopefully they can fix it soon.

Having numerous electronic gadgets and gizmos, I naturally have lots of wall wart transformers.  These things are the bane of my existence.  They occupy two or more outlets, and the more egregious ones seem designed explicitly to block two wall outlets, thereby monopolizing that particular outlet.  The worst offender, though, has to be the transformer that came with an Ikea lamp I bought a while back.  This thing is about 2” wide, 3” deep, 2” high, and weighs about 5 lbs. 

While I’m starting to see a few smaller and more friendly transformers that don’t monopolize outlets, there still seem to be lots of manufacturers who aren’t getting the message.  Which is why I highly recommend Outlet Savers (you can also get these at Fry’s).  Whenever I get a chance I pick up one or two of these and it always seems like I’m using another one somewhere.

In the long run it occurs to me that having so many wall warts is inefficient for items that remain plugged in long-term.  In the back of my mind I envision a secondary, low-voltage, distribution system built into houses to provide for these sorts of items.  There are some downsides, though, in that low-voltage DC tends to require larger wiring for large loads and there is a proliferation of voltages for the various devices. 

For portable devices I’m noticing that more and more of them are moving to USB for charging, which at least makes for a handy standard (USB specifications call for a powered port to provide up to 500mA at 5V, if I recall correctly).