Slashdot Mirror
3Com Class Action Suit
Petit-Monsieur Pas-de-Cou writes "3Com Corporation has been sued
in California by an alleged nationwide class, asserting claims relating to the advertising
of modems using 3Com's x2 modem technology. Among other claims, plaintiffs in these lawsuits
have alleged that 3Com engaged in deceptive advertising by claiming that modems employing x2
technology could achieve 56K speeds and/or were twice as fast as prior generation modems.
" Wow-that's a lotta legal statements.
I have been around modems a long time (back when I had my 300 Baud modem on my C64). There was a time when everyone said 9600 was as fast as traditional phone lines could support. But then even higher speed modems started to come into play (USR 19.2s and then they finally get a standard and 14.4s come out and start getting actually affordable). But anyone who uses modems KNOWS that you never get the theoretical top speed of a modem, with high speed one of the things that became common was error checking and bad lines means data having to be retransmited and that really slows it down. Personally I think manufactuers have done a great job continuing to push the envelope further and further. A 56K is faster than a 33.6 and a 33.6 is faster than a 28.8 but NONE of those will hardly ever achieve their actual maximum. Now we have this thing about HD manufactuers using 1000MB=1GB when selling HDs. Of course in this case the consumer stupid enough to want to sue because of their own incompetancy will also probably be too incompetant to know the difference in that case. At any rate I hope this case doesn't hold up. It's not false advertising since 56k CAN achieve 56k (on short and very clear phone lines) and that it is twice as fast as a 28.8. Maybe these people should have talked to someone who knows the technology before buying, but even then what are they going to do? It's not like they can get a faster modem from someone else.
p.s. What *really* annoys me how they decided to rate CD-ROM speeds. Talk about something arbitrary and completely silly.
I thought I read somewhere that sCSI as a more complicated controller on the drive itself, which accounts more much of the benefit. That would also make it more expensive.
But I understand the point. SCSI drives can't be THAT more more expensive (there isn't $100 worth of logic on the drive, and a scsi controller card can't cost $100 to produce than EIDE).
Buyer Beware. When 56k modems came out, I saw tons of information, that wasn't very hard to find, saying how FCC regulates the speed down, and also how connect speed depends on phone line quality, how far your are from the fully digital part of the telco network, etc, etc. The information was there, and it wasn't hard to find. I even think on the package there's an asterisk that says "FCC regulations require that blab blah blah."
My point, the information was there, all over the place, its just that people don't want to look for it. 56k is max connection speed, and I'm sure under lab conditions 3Com managed to get it up to that speed. I bet the "average consumer" won't know it if they connect at 26.4 kbps (like me) as opposed to 28.8 and that a 36.6k won't do them any good. Some of these people I bet bought a 36.6k modem even though their phone lines maxed out at 26.4k. So, when they quote a connect speed, that is a maximum, but crappy phone line noise can slow it down.
Well, ISDN is still a pretty expensive solution for not that much more bandwith, and it is a hassle to install and have to go through the phone company to install it. Compared to satellite, ADSL, cable etc ISDN isn't anything to brag about. And cmon, IDE isn't that bad. I certainly don't mind being able to add 17 gigs for less then 250. So what if it isn't the fastest for the "mission critical" apps. I am using it for data storage, and it doesn't have to blaze along. With the new IBM deskstars and other drives, they seem to be competing with scsi on storage space and sometimes speed.
As to the 56k issue, it really isn't that bad. It depends entirely on where you live, and you don't have to limit the connect always. Where I am my 56 k does fluctuate, and only sometimes receives the connect. But at a relatives they received 52000 connect every single time. It may not be the wonder it was always advertised, but as always there is a caveat. In the advertisements and brochures the fine print was there...telling people it may not work where they lived. So I don't think 3com will necessarily lose the lawsuit.
This is incorrect. The FCC limits specific levels of power output on phone systems to prevent crosstalk seepage. USR could not figure out a way to get to 56K without exceeding these limits, so they imposed a limit of 53K.
However, Rockwell and Lucent both make chipsets that are capable of reaching the full 56K because they devised a solution that didn't require exceeding the FCC limits on power output.
So you can get a true 56000 connection with v.90 or K56flex, just not with x2 or USR's v.90.
--
The world is neither black nor white nor good nor evil, only many shades of CowboyNeal.
I'll agree that 56k is a nasty hack. The problem is it had to be done. The reason it exists at all is that the local phone apparently are determined not to be dragged into the 21st century, and intend to claw the ground kicking and screaming all the way.
ISDN? In order for me to get the same deal I have now but at ISDN speeds (single channel), I would have to pay $575 a month! I do not HAVE $575 a month to spend on internet access. Without 56k modems (connecting at 41-44k), I would be at 26.4k, not ISDN. In fact, at $575/month, I could run Gigabit fiber from home to work and break even in 29 months (If I were allowed access to utility poles and conduits). As for DSL, I am promised it will be available in my area EVENTUALLY. I live in a metro area BTW.
The entire history of modems has been an end run around telcos that haven't improved in any substantial way since the mid '70s (and that was a small improvement on a system that itself hadn't changed since the '50s). If you think about it, the whole idea of analog encoding of a digital signal looks kludgy in itself. The only reason it was done was because analog circuits were widely available (POTS) and digital circuits were nowhere in sight.
As far as the class action suit goes, I don't think there was deliberate deception. I was completely unsurprised to see that 44k was the best I could get. What it amounts to is that modems are now being bought by a different class of consumers. When modem sales were restricted to people who understood computers, the probability that the modem wouldn't reach it's top speed went without saying. Apparently, it's going to have to be said now. (Warning, do not eat this modem. May cause injury if used as a frisbee, Consult your physician before using any electronic device....)
>Why doesn't anyone form a suit against these
>lawyers for misuse of the legal system?
It's kind of tough when what they're doing is perfectly legal, and they "win" by getting a settlement.
But one of my pipe dreams is a class action against the so-called trial lawyers, the personal injury plaintiffs' bar. While it is normally the plaintiffs that form a class, it is also possible to form the defendants into a class.
And the tort? Warning labels. The silly warning labels have become so prevalent that they are ignored, eliminating the value of serious warning labels. So find people who have been injured by not reading warnings that would have been useful to normal people . . .
It's not from the plaintiffss' bar, but as an example: earplugs/ear protectors are required on constructions sites near certain equipment--because the mandated sound levels of the back-up beeps exceed the permissible levels for the sound. Hmm, and since people wear the sound-blockers, maybe we need to make the signals louder . . .
hawk, esq., who thinks the warning label on McDonald's coffe should read, "Only a low grade moron would hold this cup with her upper thighs and remove the lid in a moving vehicle."
Disclaimer: I am a lawyer, but this is not legal advice. If you need advice on this matter for yourself, see an attorney licensed in your jurisdiction.
This is a fairly typical class-action settlement.
The underlying claim will strike most people familiar with the issue as silly. In all seriousness, how many people in the US took a 56k for the exact speed of 7kBytes/second? 3? 5? They diddn't buy it for a particular data rate, but for being the fastest available.
And the defense: This looks like one more that the defense would win after fighting. BUt it comes down to:
The settlement: coupons, and pay off the plaintiff lawyers. Coupons are becoming an increasingly common way of handling nuisance class action suits. There have been rumors in the past of manufacture trying to rummage up suits for this very purpose. It either locks in the consumer to buy from the same manufacturor, or it doesn't cost anything.
That is, it doesn't directly cost anything. Switching hats briefly, and speaking as an economist, this drives up prices by distorting the demand for the product. Class members pay less than they otherwise would have, but everyone else pays more. As a corrollary, the net settlement ot the consumer is less than $15, as the $15 relates to the new higher price.
Since they could win, why does the manufacturor settle? Quite simply, it's cheaper. Instead of the legal costs of fie years of litigation, the depression of the stock price from having to report the litigation in reports, and the general effects of "consumer advocates" screaming, the company cuts a bunch of coupons, and pays hush money to the plaintiff lawyers. (MY civil procedure professor referred to these suits as a great way to get paid just for going away).
The reason that class action suits exist is the notion that they are an effective way of handling suits that are too small to bring individually, and that they cut down on the required judicial resources--it's not worth suing a major corporation ofer $100.
On the other hand, when the damage per consumer is less than forty cents, it makes no sense to worry about the matter. So the system gives them a ten cent coupon, and their lawyers a couple of million.
The only class action that I know of that has actually benefitted people who were actually injured was the Iomega settlement, in which Iomega paid rebates that it had wrongfully withheld--and in full (plus a disk). TYpically, the payment to class members is negligible, or the connection to the alleged injury spurious (e.g., the breast implant litigation).
In practice, iomega excepted, the only beneficiaries of the class action system are the attorneys who feed from it.
Another thing to remember is that for 56k you need to be (if memory serves) within 5km or so from your CO. Longer than that and it's almost a certainty that you will be connecting at 33k6.
Just something else the don't tell you in the fine print.
It's either 2400 or 9600, but I believe it's 2400.
older modems (not oldest, they used discrete frequencies) used both phase angle and amplitude to encode bits to send to the other side. 9600 baud was really 2400 baud, but it sent four bits at a time. Imagine it like the old crappy parallel port cables which could do 4 bit transfers.
as modems get faster and faster, they still stick with this 2400 baud (a signalling rate) but pack more and more bits ber baud. What you end up with is what's called a constellation pattern. a 000 is sent as a 3/4 power 24 degree signal, a 010 as a 1/8 power 0 degree signal... picture a grid that's 8x8. x is your phase, y is your amplitude. whereever the lines cross, there's a possible signal.
I'm a little dozy here after lunch but I'm pretty sure that's how the terms are used. baud is the signalling rate, and bps is the signalling rate times the number of bits sent per signal.
now there is a limit as to how "fine" this grid can get. ideally you'd want infinite changes in both phase and amplitude, but you're not gonna get it. On connect, modems do all kinds of fancy frequency and amplitude sweeps and basically "shape" the channel they're in so that they can then use adaptive equilization to flatten out the frequency spectrum and get optimal communications. I think that 8-QAM is the best you can get right now.
BTW, IANACE (communications engineer)
The power restriction is there to minimize crosstalk between the various different lines between the CO and the end user (i.e. where it's still straight analog twisted pair wires bundled up together). Push to much power through, and people will end up hearing your modem hiss at the ISP underneath their call.
"'Tis great confidence in a friend to tell him your faults, greater to tell him his." --Poor Richard's Almanac
Well, ISDN is still a pretty expensive solution
Because it isn't mainstream. Also because of regulatory issues in the USA, as I understand it.
or not that much more bandwith
It's not so much the bandwidth, as the low latency, fast connects and total reliability
IDE isn't that bad. I certainly don't mind being able to add 17 gigs for less then 250.
But if IDE didn't exist you would be able to add 17 gigs of SCSI for less than 250. There's nothing inherently more expensive about SCSI, it's just that the existence of two incompatible standards has enabled the disk manufacturers to overcharge the high-enders, knowing they don't have the option of going to EIDE because limited cable lengths and the low maximum number of units would make it unusable. And the max cable lengths on EIDE are much shorter than people think. That's apart from all the other issues people have because IDE keeps running out of bits every other year.
When I read how 56k was supposed to work, I thought it the most gross hack imaginable, and it seems it is. Almost everyone has to limit the top speed in order to stop it flaking out, dropping the connection at random moments and generally being a piece of analogue technology pushed far too far.
Apparently one of the best places to put a 56k modem is on the analogue port of an ISDN adapter. That way the analogue signal has the shortest distance to travel.
I don't know about QAM, but I can explain the difference between baud and throughput.
.OO. . . ..O -OOO.
Baud is a measure of signal changes per second. Thus, a 1200 baud, 1200 bps signal might look like this:
- - -- --- - - - - - - --- -- -
(where '-' is on and ' ' is off...it's a graph versus time)
Well then someone found out that it was practical to carry more than one bit on one signal change. Hence, here's a 1200 baud, 2400 bps signal (2 bits per signal change):
. -O -O - -.. O - .
(where ' ', '.', 'O' and '-' are all distinct signals)
The problem is that, especially with poor quality phone lines, it's hard for the modem to distinguish between the different signals. "Is 3V closer to 5V or 1V?"
As for the actual maximum number of baud (signal changes per second), I'm not sure, but 9600 sounds about right. For a 36.6 modem, that means 4 bits per signal change?
Its $72 a month when I finally dropped it here in Starkville, Mississippi. Sure, the 15KBps transfer speeds were nice, but with a 56K modem, I hardly know the difference with Netscape. I now pay $25 a month, not the $100 with ISDN (taxes, etc...)
The real reason I dropped ISDN is that the connection would die about every month or two and getting to a like human on BellSouth's "business" repair line was insane. Once I logged 6 hours on my cell phone to finally to convince them it wasn't my problem and get a tech to get the line fixed. Try getting the runaround and telling every supervisor and department that you put a scope and a meter on the line to prove it was an opened circuit (10 megaohms) with no signal. It was a game I got tired of playing.
Its analog for me at the moment.
I've been looking at digital cameras, hoping that they will soon become a viable alternative to 35mm photography. I think these people are some of the worst offenders on stretching specs.
Most egregiously, they count the RGB sensors as three separate pixels. By this logic, the laptop I'm typing on has 3072 x 768 pixels; pretty impressive, hm? On the very best of the digital cameras (the Nikon 950), the effective number of pixels (measured using resolution targets) is about half the claimed spec, and even then the image suffers from moire, chromatic aberration, and other artifacts.
What really opened my eyes was looking at Sound Vision's camera, which is only 800 x 600 pixels, but blows away the "megapixel" cameras in total image quality. Unfortunately, because this camera has to take three separate shots through three separate filters (R, G, and B), it's only useful for a limited range of work, so it's foundering.
As if this weren't bad enough, Kodak came out with a digital camera with about a million "pixels", but marketed it as about 1.5M, because they estimated that it delivered quality slightly superior to the 1.3Mpix cameras they were competing with. Can you imagine printers competing on this basis? At least dpi claims have tended to be fairly solid (although I'm more than a bit suspicious over Epson's 1440).
Caveat emptor, I guess.
LILO boot: linux init=/usr/bin/emacs
what the hell is wrong with the people who are complaining...
I have gotten 53k connects, and I have gotten as low as the high 40s... These people who aren't getting better than 24000 are probably using category 3 wiring for a quarter mile in their house, exposing it to all forms of radiation and crosstalk...
Beyond that - It says on the box that you won't get 56k... It also says that you may not even get the 53k if your line is not prestine... Maybe they should have read the box before purchasing.
At my last home I had beautiful self wired phone lines, and I was less than 6 blocks from the central office... 53k was absolute...
Where I live now the phone line is shared with my neighbor and spliced quite a number of times... While standard 33.6k modems will only connect me at 26.4 -31.2 tops on this line, 56k modems get me up to 44k... I am quite greatful...
Ofcourse I did order new phone lines, and now I am not sharing and am back up to 53k, so 44k can kiss my ass...
thanks you USR and all other parties involved, nice work...
--
Marques Johansson
displague@linuxfan.com
Marques Johansson
I think the bug is in the modem firmware, and not Windows (or other OSs that report 115Kbps). If I recall correctly, the 'Hayes compatible' way is to report the modem speed (DCE) where some modems report DTE without an init switch.
Occassionally you will run into someone that has actually switched modems because Modem A reports 19.2Kbps, while Modem B reports 115.
--
Business. Numbers. Money. People. Computer World.
You have to be careful with how you initialize the modem. The initialization string I was using for a few months actually somehow disabled high speed connections. When I trimmed it down, things started working. And there are locations from where you can't connect--too noisy or too far from the switch. But my phone wires aren't perfect (water damage), so there seems to be at least some robustness.
My 56k modem came with my PC. I was quite sceptical about the 56k stuff and probably wouldn't have bought it, but I'm happy with it now. It will still be months until DSL becomes available, and ISDN is much more expensive.
Of all the calls I get here in my company's call center, the whinney, cry-babies who can't get 56K on their phones are the worst.
It's especially funny when they call in on a line that is so noisey I can barely understand them through the static. "Why can't I get 56K speeds? I can only connect at 19200!".
Sometimes they have 20 feet of phone cord between their computer and the wall when their computer is right up against the wall and 2 feet from the walljack... I could go on, but why waste even more space...
-- 100% MS-Free as of 4-4-1999, 11:47:38 PST. "The lapdance is always better when the stripper is cryin'" Free Kevin,
Let me clarify on point 2 a bit. Analog phone lines (commonly called POTS, or Plain Old Telephone Service) have a frequency response (if I remember this correctly) of about 40 Hz to about 4000 Hz. This means that *all* of the equipment between one phone in your home and the phone that you are calling in someone else's home, has to handle that frequency response (or a reasonable approximation to it). The telephone switches are capable of sampling that frequency response, digitizing it, switching it, trunking it, etc.
:/ 3Com's modems do x2/v.90 quite well actually...please remember the adoption phase of v.34...it progressed at about the same speed...particularly with new modems, this stuff is done really in software (DSP's running software that perform the modem functionality)...particularly for central site modem equipment (with 3Com HiPer DSP's for example, they use a single PPC chip to be 24 modems, plus other functionality)...and we're still basically dealing with version 1.0 of this v.90 modem software, well, maybe we're at version 1.1 at this point...compatibility is getting better.
:)
ISDN is a bit different...its a digital connection to the home, so the terminal adaptor or ISDN router is responsible for digitizing the data...and for analog signals on the back of your TA, it does it in much the same way that the telco switches do for the analog line above...for a data connection...well, the data is already digital from the computer, so its just a matter of encoding it for the line protocol that ISDN uses...thus it soak up the full 64Kbps channel available to it. In the telco switches, its just a matter of taking that already encoded data and switching it...this is fairly trivial since you're already dealing with digital data, its just a matter of taking it off of one line and throwing it on another (well...there's more to it than that in implementation of course, but the basic idea is simple). However, because the telco network was set up with 64Kbps (sometimes just 56Kbps) channels for transmission of the digitized audio data, ISDN limits at 64Kbps as well...this is a direct result of the analog connections in the first point. The whole telephone system was set up to handle 64Kbps chunks, so that's what ISDN gets, since its transmitted across largely the same telco infrastructure.
xDSL is a whole different beast entirely. As I figure most of you are aware, the physical copper wires that are run into your house (business, whatever) are physically capable of handling a much greater bandwidth response than the telco's are making use of (*this* is where Shannon's limit really comes in). So, xDSL uses frequency responses that the telephone network has never made use of before (4000Hz up to some really bignum Hz). However, because the telco infrastructure isn't designed to handle these frequency responses (and bandwidth quantities!) the switches and trunks and other telco stuff can't handle DSL directly... Here's where the really big limitation with DSL really is...the DSL equipment has to be on the copper line *before* it hits the switch, or, if the signal goes through the switch, it gets clipped back to the 40-4000Hz signal that the switches are designed to handle. So, your DSL equipment has to be at the telco Central Office (a considerable expense for non-telco's to co-locate equipment), your phone line (actual copper wire) has to be rewired to hit the DSL equipment and then split off from there into the switch, and because the switches can't handle the frequency response (and bandwidth) it can't be "switched" meaning you can't place calls on DSL...really...in the telco world, this seriously limits the usefulness of DSL. Even in the ISP world its somewhat limiting...if you are using one ISP for your DSL, you can't hang up and dial another DSL provider...you'd have to have a completely seperate DSL line in from that second provider to use that...kinda sucks.
Don't get me wrong...DSL is really cool...but there are some serious limitations.
To get back on topic a bit
Oh, another point...before you go off and complaining to your ISP about your v.90 connect speeds, the critical side of a v.90 connect is the client side modem...ie, the one attached to your computer, not theirs. For example, if you have a Lucent WinModem, you better make *darn* sure you have at least version 5.32 of the code on your modem before you start calling up and complaining to your ISP or you're going to be sending them some sheepish apologies eventually. There are other code revisions that help with other types of modems...a good resourse is http://www.56k.com for finding this information.
Sorry so long.
Jeff
2nd, with all digital technologies like xDSL, and ISDN, you get much higher data rates because you never have to convert the signal. POTS, ISDN and xDSL all use the same 2 wires, but with various encoding methods, and consequentially different equipment both at your end, and at the telco end. The biggest reason why DSL is so much faster than ISDN is because the xDSL spec has shorter distance limits that ISDN does. It's pretty similar to the distance restrictions of Gigabit/Fast/Ethernet.
The reason that you hear lots of people complain abot the FCC and 56K, is not that they want to restrict bandwidth, but that they have a limit on how much power you can send across phone lines. Because of this restriction, the tricks that the 56K modems use to get above 33.6 can't be maximized on "crystal clear" phone lines. You'ld be hard pressed to find a line where it mattered anyway, but that's another issue.
56K modems are a pretty neat trick, and most people I know do have 56K. I decided about two years ago to just hold out for one of the current multi-migabit technologies. I'd spent too much already on 14.4, 28.8 and 33.6 to spend a dime on any new modem. I pondered ISDN for a while, but it is _so_ much money. I just can't justify that monthly bill.
here ends the lesson...
-earl
This is all actually kindof funny. According to the not-quite-finalized agreement, if you bought a modem that's in the set in question, then you are entitled to a "$15 Rebate Coupon." Not $15, mind you. A $15 coupon that's good only for purchasing more 3Com products. If it didn't have the phrase "Class-Action Suit" attached, I'd think it was just another promotional campaign.
I was personally glad that 56K modems came out. My transfer rates jumped from 3.5kps to 5.1kps when I went from 33.6 to 56K and a 50% speed increase may not be double, but it certainly is enough to be noticable.
I am on a cable modem now and I could never go back to a plain modem, but for people who are in areas where they can't get highspeed access, they're better off with 56k than 33.6.
Just my 2 cents.
Just a SkiBum stuck in the east...
In days past, if you bought a 300 bps modem, you would expect nothing less than 300. This probably held true up to about 9600 bps, then, gradually, we got used to not connecting at full speeds. It's funny now that the happy posters in this topic are happy at 45K, not 56K or even 53K. Nobody would have been happy with 220 bps from their 300 bps modem.
But modem's aren't special; they're just following the trend. Show me a printer which will really live up to its pages-per-minute spec. In fact, show me an old dot matrix printer that got anywhere near its characters per second spec. Show me a monitor that's actually sold based on its viewable screen area. (Actual viewable area is only a footnote now, and only because of another one of these class action suits.) And I'd take any hard drive or CD-ROM drive claims with a huge grain of salt, too.
As consumers, all we can do is become informed. I don't expect the manufacturers to start selling 17.9" monitors -- they'd be afraid of being excluded from a PC Rag's roundup of 300 19" monitors. But maybe we can insist on more realistic reporting. It just might be possible now that there are lots of web sites reviewing hardware and basically making PC Rags obsolete.
They lost this suit a while back. This is a requirement of the court for them to notify all their customers about the refund. I notice they haven't take out any full page ads in the magazines where they normally advertise.
:-)
Just go back and RTFLD, its in legal-speak so you have to read it sideways, preferably after taking some mind-altering substances
IANAL, and I Hate Lawyers
the AntiCypher
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on
I have my own phone switch, and a pile of different modems for testing, and various access servers for terminating the 56k end.
:-) (actually, we use our poor little switch for purposes that would be illegal and dangerous on any publicly connected switch, but thats another post)
We use it to prove that different modems only connect at far less than the 53.3k even under the most ideal conditions. We measure the length of the local loop in inches
But USR modems are no worse than any of the others, even at ideal conditions. Sometimes they hit the theoretical 53.3 max, sometimes they don't.
Me thinks this to be a lawsuit against all the advertising and slimy sales promotions USR is known for. Certainly their technology is about the same as all their competition.
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on