February 09, 2007

Moore's Law cut in hafnium?

Pardon the pun, but I'm still in shock that I missed this news. Last week, Intel announced that it is switching away from silicon for its next generation of chips and moving towards metal and a "high-K" hafnium compound. This should let them shrink down to 45nm gates from the current 65nm gates as well as improve the power efficiency.

From New Scientist Tech:

A transistor consists of an electrode that switches the current on and off within a "channel" using an electric field. In the past, to make the transistor switch faster, and thereby up its performance, chip makers shortened the electrode and thinned the insulating wall that separates it from the channel.

This is far from ideal, as thinning the wall causes current to leak from the channel into the electrode, wasting heat and electricity. Furthermore, it means more current leakage than the transistor could handle.

Now, in an effort to continue shrinking and speeding up its transistors, Intel has come up with an insulator that transmits a fast-switching electric field even at a relatively large size. The exact composition of this "high-k" material is a secret, but Intel says that it contains hafnium. It is claimed to increase transistor switching speed by 20% and leak five times less current.

In 2003, Intel also had to tweak its process to start making 90 nanometre transistors. Its secret then was to use "strained silicon" in its transistors. This increased the speed at which current flowed, although Hutcheson says that advance was "a walk in the park" compared with achieving today's leap to high-k insulators.

The change in insulator has also led to a change in the gate electrode material. When high-k materials are deposited next to an electrode made of polysilicon, defects normally arise at the boundary. But this effect disappears when a metal gate is used instead.

This is not just a research project. It's going into manufacturing later this year under the code name "Penryn" and will be available in products in 2008. Specifically, look for the new versions of the Core 2 Duo, Core 2 Quad, and the Xeon processors next year.

In short, this should extend the lifetime of Moore's Law well into the 2020's. Intel and AMD are rumored to be close to a similar announcement as well.

Technology by Dan | Permalink | Comments (0)

April 13, 2006

Dictaphone

Lately I've been trying to get more organized. It's been an uphill battle, especially with various debris crashing down on me in the occasional avalanche, but I'm making some progress.

One of the things that I was battling was the "think of it in the shower" syndrome. That's when you remember something important but only at times when you can't do anything about it, i.e. remembering in the shower that you need to call Jim. (Yes, Jim, I'm thinking of you constantly in the shower -- just so you know.) The key is to capture those thoughts somewhere in a trusted system and then go over them later when you can organize them into actual actionable to-do items.

I do have a PDA, but I've found that it's not quite convenient enough. It's a little too big to carry around in my pocket, and even then, writing something down isn't always practical. I'd tried carrying around a little tape-recorder before, but it suffered from the twin problems of size and tape management. It was about as big as my PDA, and the resulting recording was a long rambling monologue that had to be played, paused, played, paused, etc. for transcription.

What I wanted was something very small, i.e. smaller than my cell phone, that could record at a touch and organize those recordings into discrete chunks that I could manage independantly of each other, preferably on my computer.

Well, it looks like I found it.

dictaphone.jpg

It's actually an MP3 player, but it also functions as an AM/FM receiver and a dictaphone. It remembers what mode it was last in, so I can put it into dictaphone mode and just leave it that way. It's not quite one-touch, but it's close. I have to turn it on, hit a button to start recording, hit it again to stop, and then turn the thing off. Still, it's reasonably simple to do one handed. The resulting recordings are encoded to MP3 automatically and numbered. It connects to my computer via USB, and it opens up as a file folder showing me each recording. I can play them one at a time, act on them, delete them, or copy them to my computer for archiving and/or later action. It's pretty much exactly what I wanted, and I picked it up for about $70 at Fry's Electronics.

Now, if only it was waterproof I might finally get around to calling Jim.

Technology by Dan | Permalink | Comments (0)

October 20, 2005

Orgasmic Technology

It's been a while since I've made a technology post, but this new Bluetooth add-on is just too good to pass up on.

And by the way, I've been thinking lately that it's time to upgrade MAW's and my phone plan to include SMS text messaging, and maybe Bluetooth phones as well. This is, uh... a complete coincidence.

Thanks to nukewolf for the tip.

Technology by Dan | Permalink | Comments (0)

August 25, 2005

I am officially unique... and evil!

After grinding through the legal mill for about four and half years, my first two patents were awarded today. Hence, I am officially unique... twice. And for those of you anti-IP folks out there, I'm officially EVIL!!

(Note: my company is historically against software patents and only started getting them after being sued on an infringement of such. Since then, they have a history of only using these patents defensively, cross-licensing to gain access to other patents. So, I'm not really that evil.)

My only real regret is that it didn't happen two weeks ago. That's something my father would have liked.

Technology by Dan | Permalink | Comments (1)

June 03, 2005

Lost and Found on the Internet

I just found a new service that I wanted to share. Basically, it's a lost and found service for the internet. StuffBak will give you small stickers (or perhaps you have to buy them...not sure) that you can place on various items. It contains their URL, an 800 number, a unique ID, and in bright yellow letters "REWARD FOR RETURN!"

The idea is that if an honest person comes across your $400 PDA or camera or whatever, they would like to return it to you. Even if they're a little less than honest, a reward might induce them, but they still have no idea of how to find you. Well, the StuffBak sticker gives them a way to report the item, and StuffBak will arrange the shipping and return to you. The service is technically free until they actually return an item to you, and then it's just $14.95 plus shipping (typically less than $30 total).

Yes, you could achieve much of this just by engraving your contact info onto the item (if it's big enough), but there is a nice convenience and privacy factor here. I have a similar service for my keys, provided by my alumni association, where I carry a keychain ornament that allows a finder to just drop the keys into mailbox, eventually getting them back to me for a fee.

Now, I did say it's "technically" free. It's really just free for the first year after the activation of the label. After that, it's either $2.95 per year per label, or $20 for any number of labels for five years. (You do the math.) Also, any reward you promise is obviously paid by you, but StuffBak handles the details, providing immediate gratification to your hopeful Samaritan.

A field test run by CNBC showed that 5 out of 6 items were returned within three days, so it seems to work. (Plus, the one item not returned was so cheap that I wouldn't have used the service on it in the first place.)

Anyway, I just thought I'd pass it on in recognition of, "Hey... that's a good idea."

Technology by Dan | Permalink | Comments (0)

May 19, 2005

New Stem Cell Technique

Well, while we're on the subject of clones (insert Star Wars joke here)...

South Korean scientists have managed to create functional stem cells directly from adult patients, not using a fertilization or embryonic middle step.

...their method may be less controversial than other work with embryonic stem cells because, by their definition, a human embryo was never actually created.
...
Hwang said his method differs from that first used to derive human embryonic stem cells in 1998 and he proposes using a new term for the cloned embryos -- a "nuclear transfer construct."

"I think this construct is not an embryo," he said. "There is no fertilization in our process. We use nuclear transfer technology. I can say this result is not an embryo but a nuclear transfer construct."

I have some hope that this development will clear up much of the current stem-cell controversy since it should side-step the whole pseudo-abortion issue. You could make the argument here that this is no more "life creating" or "life destroying" than the common practice of donating blood in anticipation of surgery when you will need it back.

Now, unlike what many Democrats have promoted, I don't think that miraculous cures for paralysis and organ failure are right around the corner, only lacking the application of federal funds, but I have high hopes for some good stuff in the fifteen to thirty year timeframe.

Mostly, I suspect that the cure for cancer might some out of this. File this one under "Things we believe but cannot prove", but I've long suspected that cancer is caused by a cell being trigger back into stem-cell mode by accident but without the proper chemical instructions of what to do. If we ever learn to control stems cells the way the optimists promise (e.g. regrowing nerves or organs in place), then we should learn enough to thwart stem cells gone bad.

My father is dying of cancer right now. He's been in stage four for about two years, and I'll be surprised if he lives out the summer. All of this is decades late for him, but it makes it very real to me.

Technology by Dan | Permalink | Comments (0)

January 13, 2005

Annoying chime...

I use MSN Messenger a fair amount for my work, but I'm typically logged into it not from my development system, but another one a few feet over. As a result, the little message pop-up doesn't catch my eye, and the default "New Message" sound is so innocuous that I often miss it amidst the constant chimes of spam arriving.

So I set out to create a new chime that would be so annoying as to definitely get my attention. My initial attempt was so extremely annoying that I had to tone it down some. Even now, it's pretty jarring. Give it a listen!

Anyone want to start a pool on how long before it drives me insane? And no, I won't post my IM contact to make it easier to affect the results.

Technology by Dan | Permalink | Comments (0)

December 13, 2004

Timewarp!

home_computer2004.jpg

In case you're having a hard time with the caption:

Scientists from the RAND Corporation have created this model to illustrate how a "home computer" could look like in the year 2004. However the needed technology will not be economically feasible for the average home. Also the scientists readily admit that the computer will require not yet invented technology to actually work, but 50 years from now scientific progress is expected to solve these problems. With teletype interface and the Fortran language, the computer will be easy to use.

Technology by Dan | Permalink | Comments (2)

December 11, 2004

ISP Rant!

MUST... NOT... NUKE...TECH SUPPORT...

I had a little trouble with my internet connection today.

What is it about calling up on tech support for your internet connection that implicitly states, "I am a complete idiot and will lie to you about the details of my problem"? I'd like to know, because everytime I call in, that's the kind of treatment I get.

What makes matters worse is that the connectivity in my area is spotty. I get great signal strength, so that's not the problem. Just somewhere up the line things get flakey. So, I get connection loss a lot, maybe once a week. Usually all I have to do is reset the cable modem, and then everything is fine, but about once a month, I have to call up tech support.

And that's when the trouble starts. For starters, today I had to wait on hold for about twenty minutes. Every couple of minutes a recording popped up explaining to me that if I was having conflicts between Microsoft Outlook and Norton AntiVirus where you were unable to access certain mail messages, I could follow these simple directions for disabling virus checking on my email. Oh yeah, now that's a good idea.

Eventually, I got through to a human being. At least I think it was. Considering the level of help I actually got, it's possible it was a good Eliza program. I explained that I had already reset the cable modem, and of course, she asked me to reset my cable modem. See what I mean about assuming I must be lying to them?

Well, I was prepared to humor her, so I did it anyway. When it came up again with the same fault showing on the "cable" light, I explained to her that this kind of thing happens all the time, and that it's not a problem at my end but a problem further up the line. She proceeded to query me on just how I had reset the modem. Once she was satisfied with my explanation, she asked me what IP address I was being assigned.

IP address? Like, you know, the thing I would get if I didn't have a fault on the cable line? Again, I humored her. I logged into my router and read off the IP address.

"That's not one of our addresses. Are you sure you're reading the right one?"

"Yes, I'm sure, and I know it's not one of your addresses. DHCP doesn't work if the underlying packet transport if failing."

I point out again that this kind of thing happens frequently, and I ask if she could please check to see if there was a problem in the area. She says she hadn't heard of any problems. I suggest that I might have just been the first to call in. She puts me on hold for a moment before telling me she isn't aware of any problems in my area. I even offer to go to the neighbors to see if their connection was also down, but she demurrs.

Then she's wondering if I have the signal strength, and I point out that the installer and one other guy who had to come out on a service call both went on and on about the incredibly high signal strength I had. Fearing problems in this, I had even told the installer guy to bring an amplification unit, which we chose not to install as it would be overkill.

Then she starts asking about my wiring. Now, remember, it's not like I was a way for six months and came back to find my Internet disconnected. No, I had been using it actively and then *POOF* it was gone. Nothing happened at that moment. No lightning strikes, no meteor strikes, no upper atmosphere EMP's. I pointed out how astronomically unlikely it was for my wiring to have suddenly failed for no reason, compared to the near constant problem of their network fouling up. Mind you, I was still being polite, just pointing out the mathematics of the situation. MAW was there, but she'll probably talk about the edge in my voice and how the veins in my forehead were about to burst.

Undeterred by my amazing recitation of history, logic, and probabilities, my tormentor helpful technician asked if there were any splitters upstream from my modem. (Clearly she was still after the signal strength issue.) I admitted that yes, there was one AND ONLY ONE splitter involved. She asked if I could try to bypass the splitter.

Now, let me say a little something about the wiring in my house, because I am both extremely proud of it and extremely frustrated with it. My goal was to have everything setup in a single closet dedicated to the wiring. Every telecommunications wire, from phone to cable to network, would be a single, direct run from the closet to its source or destination. There would be no connections hidden away in walls or attics. Nothing would ever "get loose" in an inaccessible spot. I even put extra ventilation in that closet so that I could host servers in there without worrying about the heat problem.

And I pretty much got what I wanted. The only problem was that the assholes lazy dipshits incompetant boobs wiring technicians who installed it did not follow my instructions on how to terminate everything in the wiring closet. I told them that I wanted it all layed out on a breadboard so that it was all easily accessible and that most changes could be made by the use of patch cables. Well, they must have figured that I wouldn't know the difference or that "no one does that kind of thing in a house", so they just dumped it all into a cramped metal box and just bundled up the connections wire-to-wire, cable-to-cable, with it all hanging in there loose.

So, when the ever-helpful technician from COX wanted me to bypass the splitter, it involved squeezing into the closet, opening up this metal box, sorting through the bundle of wires and connections, and identifying the one splitter and its input cable. Now, if it weren't for the fact that I desparately needed the internet connection back for some work I was doing "at the office", I would have just hung up and waited for an hour to see if things had sorted themselves out, but I needed it. So I got in there, worked it all out and connected the main source cable directly into the modem and reset it, and...

Drumroll please....

And there was still a fault on the cable light. I assured her that there was no equipment between my modem and their equipment outside the house. Well, now she admits that maybe it might be their problem and wants to schedule a truck to come out. Hopefully, she says, she can schedule it for today. I bite my tongue because their actual field technicians have generally proven to be very savvy, and I knew that one of them would be able to spot the problem immediately and fix it.

But first, the technician notes that I don't have "COX Wiring Insurance" on my account and that any problems the field guy finds in the wiring in my home will cost money to fix. Now, maybe this insurance is a decent idea if the cable was retrofitted into a fifty-year-old house with water damage, but in a two-year-old house with overengineered wiring and an owner who knows what he's doing, it makes no sense at all. It's just a $3/month extortion that will never pay for itself. Still... "Before I schedule the technician, would you like to add this insurance to your account?"

Ok, go with me here. She's still pushing the theory that the problem is in my wiring, but if she really believed that, then why on Earth is she offering to let me convert a $400 wiring charge into a single month's $3 insurance premium? That's like offering a good deal on life insurance to an 90-year old with lung cancer and a heart condition!! But I digress.

"No thank you. I am fairly confident that the problem is not in the wiring for my house. If it is, I will accept that risk."

So now, after about an hour from the time I first picked up the phone, she starts writing up the trouble ticket to schedule a field technician. I'm on hold for a few minutes, still cramped into the wiring closet, not wanting to move lest I lose track of which cable is which. Eventually she comes back on and says that she's forwarded it on to the supervisor and will know shortly when the field technician will be scheduled. I tell her I'm putting the splitter back into the arrangement, and I get started. She puts me back on hold.

I've just gotten the last coax on and tightened -- not just finger-tightened but wrench-tightened -- when she comes back on.

"Do you have the schedule yet?"

"I just got back the note from my supervisor. He'd like you to try to reset the modem again."

Now, on the one hand, I wanted to reach through the phone and strangle her (difficult on a cordless), but on the other hand, I had already unplugged the modem during the most recent cable switch. So I plugged it back in, and... and...

A GREEN CABLE LIGHT!! Halelujah! Halelujah! Halelujah!

For a split second I wondered if I'd been wrong about her. Maybe the problem was in my wiring after all? Had one of the cables been loose, even though I needed the wrench to disconnect them? Had my act of restringing the cables solidified a connection that had just recently become flakey? Was I, in fact, an idiot who had been unknowingly lying to her the entire time?

But the moment passed. "Tell me, just what did your supervisor say?"

"He said he'd checked something and wanted you to 'try it now'."

"He changed something?"

"Yes, I think so. There might have been a problem in your area."

"A problem in my area." The words echoed from half an hour ago. "Well, it's working now."

I spotchecked a couple of computers to confirm that the packets were flowing and started wrapping things up. Part of me wanted to take her back through the steps, pointing out that I had known from the very beginning that the problem was on their side, and that I had told her this repetitively, but that she had ignored me. I wanted to show her how she had wasted both my time and hers, when all she really needed to do was write up the ticket and have her supervisor push whatever magic button he'd pushed to fix it. Mostly, I suppose I wanted an apology for having been treated like an idiot who would lie about the problem in some perverse scheme to get it fixed faster.

But I knew I wasn't going to get it, so I just thanked her for her time and ended the call. Closing up the wiring box was another ordeal, one which has me very close to hiring a professional wiring guy to come redo that closet the way it was supposed to be done. And then I went back to my work.

Now, maybe the other 99% of their callers are idiots, and they have to take those precautions to avoid dispatching repair vans out to someone who will say, "But why should the cable be connected? My laptop has a wireless network card." But I'm not that guy. I'm the guy who knows more about the network than you. I'm the guy who might even know more about the network than your boss. I'm the guy who should get to call in on the Red Phone.

But there is no red phone.

So I sit here in dread, awaiting the next time I have to call, knowing that the first thing they will say is, "I understand, sir, but could you try resetting it again?"

Narrative /Technology by Dan | Permalink | Comments (3)

October 31, 2004

Electronic Voting: reiterating the obvious

We’re heading into our first presidential election with widespread use of electronic voting machines. This has been filled with controversy, both because they haven’t been deployed as widely as some wanted and because many feel that they can’t be trusted as far as Kerry can throw a Silver Star. The irony isn’t that we couldn’t work out this problem in the four years since 2000. The irony is that the solution was obvious, widely known, and largely ignored.

One of the reasons the new electronic voting machines haven’t been rolled out to that many places was money. They are, after all, not free, but there was enough popular support for some kind of improvement that the money would have been there at the federal level. Around 2000, I’d heard numbers of two to four billion dollars being talked about to revamp the voting system nationwide. That would have made a budget of a few thousand dollars per voting machine and still had enough for a massive nationwide rollout. So, I can’t really blame the problem on money.

Instead, it’s about certification, and that’s about trust. While federal dollars could have paid for it, the running of the actual election, i.e. the physical process of taking the votes and processing the ballots, is largely a matter of state law. Thus, it was up to each state to certify any electronic voting machine prior to use. This seemed to be going all right until it was pointed out by some techies that the process was insecure and ripe for tampering. I happen to agree with them, and I am a techie. (A long standing irony is that it’s the folks who understand the technology the most, trust it the least – because they know how fragile it is.)

Why aren’t the machines trusted? Why are they so insecure? After all, the guys who did this are professionals. Some had industry experience in making ATM’s, an area where security literally means dollars. I can only explain it as a mixture of stupidity and laziness, because it would seem that the answer to the security problem is obvious. Of course, I also blame the federal and state governments. The solution should have been obvious to them as well, and they (particularly the fed’s) should have put out a specification, a set of requirements, for any electronic voting system, and those requirements should have ensured a high level of trust. But they didn’t. Instead, they just got out the checkbook and asked what was available.

So, what is the obvious solution? I’ve been keeping you in suspense this long, but you won’t be surprised. After all, it is the obvious solution: paper backups. That’s right, annoying, troublesome, chad-hanging paper. However, this inclusion of paper seems to be an anathema to the electronic voting companies – a surprise given that ATM’s have to produce paper receipts and have done so for over twenty years. Some even say that paper is an invitation to even more fraud, but I just don’t see it. Paper is much harder to forge than bits. It’s physical, and it’s spread out in a lot of places.

So here’s Dan’s fraud-resistant voting system. I’d say fraud-proof, but I can’t be certain. Still, it’s pretty good.

Step 1: Voter steps into the booth. She votes on the electronic screen, using some input device, i.e. touch-screen, wheel, mouse, keyboard, etc.

Step 2: Machine prints a receipt (image below). The user looks at the receipt and the screen and confirms that they match. If there’s a problem, or if she just changed her mind, she can call over the polling official who will enter a code to reset the machine and also destroy the receipt – then go back to step 1.

Step 3: Once the voter is satisfied that the machine and the receipt are in agreement, she hits enter on the machine and puts her receipt into the traditional, locked ballot box.

Step 4: At the end of the day, the votes are uploaded from the machines, moving up the precinct chain to the county and state levels where winners are determined promptly.

Step 5: The day after the election, the paper receipts are tallied by electronic readers (details below). This happens for every election, not just if it’s close, not just if someone asks for a recount. It happens no matter what. If there was massive fraud on the purely electronic version, it will show up immediately.

Step 6: A small percentage of paper receipts are chosen at random to be inspected by hand to make sure that the electronic readers are reading them accurately. If there was massive and matching fraud on the readers, it should show up in this manual inspection.

Step 7: Any major or minor discrepancies between electronic and paper totals can be investigated to look for either fraud or simple mechanical, electronic, or systemic failure.

Step 8: Once any discrepancies have been cleared up or declared trivial, the appropriate state or county official will certify the election results. This is a process that occurs even now, and it’s this action that makes the results official, not the initial reports on election night.

Since I’m asking you for so much faith in this piece of paper, let’s look at it:

eballot.gif

First of all, note that it is in all-caps optical-character-recognition (OCR) font. Given the redundancy noted below, reading this should be very reliable, or at least, it will be highly unlikely for an error to slip through.

It should be printed on either common 8.5 x 11 paper or a common roll-paper. The idea is that if a machine runs out of paper due to heavy turnout, polling officials should be able to purchase replacement at any office supply store.

The top line records the information necessary to track the vote down to the precinct polling station and the individual machine. Thus, any cross checks between paper and electronic can track it down to a specific place and device.

The second line uniquely identifies the individual vote cast by both a number and a time stamp. Those two should proceed forward in tandem.

Next comes the votes themselves, and each vote comes in four columns, which are actually redundant pairs. The first two columns identify the choice being made, both by number on the ballot and by human-readable name. If they don’t match, something is amiss. The second two columns are the choice made, both by number on the ballot and by human-readable name. Again, if they don’t match, something is amiss.

At the bottom is a long string of hexadecimal numbers. This is a digital signature of the vote that was cast. Specifically, it is a signature of the text above it on the receipt, signed with a key that is unique to that voting machine. If the signature is wrong, then something is amiss. In addition to authenticating the paper receipt, this should act as an excellent error detection code.

Combine all of those paper features in the error checking, and discrepancies should be easy to spot and easy to track down. If there is a major discrepancy between the paper and the electronic version, enough to swing the election, then the paper must take precedence. It is harder to forge since it contains both the physical properties (physical existence, paper consistency, printing characteristics, etc.) and the mathematical properties (record of unique instance, digital signature, etc.) that make things hard to forge. And of course, if the computers do become suspect, the paper versions contain enough information to be counted by hand by human beings who are readily accountable to laws governing election fraud.

But ultimately, the best part of the error detection occurs back at step 2, when the voter confirms that the paper vote represents what she voted for electronically. You’ll have about 110 million motivated fact-checkers making sure that the system is working correctly. Everything after that is just to make sure that error-check is honored.

The ironic, nay, the sad part of all this is that I banged out the specifics here in about ninety minutes. A good team of ten or twenty technical folks could work out the software and hardware in a year. But here we sit, almost four years after the fiasco of Florida 2000, and to my knowledge, none of the electronic voting systems are as resistant to fraud or error as the one I’ve just described. I may be smart, but come on guys, I’m not that smart.

Politics /Technology by Dan | Permalink | Comments (3)

October 11, 2004

Web Hosting: Good, Bad, and Oddly

I have recently been moving some of my domains from one web host to another, and I want to reward and punish these hosts with a little publicity. Plus, I want to share odd bits I discovered when checking out my usage statistics.

IDCSoft.com has always hosted this blog, and now hosts two of my other domains. (Why so many domains? One is primarily for email with a nice short name@domain string, and the other is for web experimentation before making something go live elsewhere.) Anyway, IDCSoft is cheap ($5/month minus a volume discount) and so far, very reliable. I've had barely a hiccup, and their support staff has been very helpful in helping get things setup. Thanks to Marvin for pointing me their way.

ADDR.COM on the other hand, SUCKS DONKEY DICKS!! Ok, they were decent at first, though they were pricier at $10/month. The big problem turned out to be email reliability. My whole reason for getting my own email domain was so that I could finally be in control of my email once and for all. I wouldn't have to change the address at the whim of a corporate merger. I wouldn't have to be dan29829284@everyone.com. And most importantly, my email accessibility wouldn't be at the whim of some flunkie in the NOC.

I actually ran my own server for that reason on my own DSL line, and everything was grand, but then I lost my fixed-IP arrangement in 2002, so I moved to ADDR.COM. There were no real problems for the first several months, but then I started getting this persistant error about not being able to log in to get my mail. Working with their support, I discovered it was because they were letting the drive on their mail server get full. First of all, disks are cheap, so buy more. Second, everytime this happened, I would email them (from work, my only reliable option in that case) to tell them, and they would fix it. But I mean, come on, why should I have to be their trouble-detector? After it happened a few times, you'd think they'd set up an automated detection system, perhaps one that even spotted the disk space getting low ahead of time.

But I could live with this kind of low-level incompetance if it saved me the day-to-day hassles of running my own server. The real deal-breaker was when they started installing spam filters on the server. Spam doesn't really affect me all that much since I have been using SpamBayes for over a year. I get a few hundred spams every day, but they all go right into my spam folder with an extraordinarily high accuracy. What these bozos at ADDR.COM did instead was to just install an black-list. Suddenly, MAW was being told that mail from her mailing lists was bouncing or just disappearing. We tracked it down to these server-side spam filters.

When we complained, they said that they were doing these filters to protect us, their customers. Well, we can protect ourselves better than they can, so we told them to shut it off. Alas, they couldn't, but we could whitelist any address by sending email to it through their SMTP server. That would be fine, except for the fact that we couldn't send out through their SMTP server. We tracked it down to the point where we could see we were contacting their server, but that they were refusing us, even though we were using the right password. They could see the problem on their end but couldn't fix it immediately. Instead, they added to the whitelist manually for us.

This held as an uneasy truce over the summer, but when I was setting up this site, I certainly didn't want to add one more site to my problems with ADDR.COM. I hadn't gotten to the point of moving the exising domains, but I was at least open to the idea. And then, last month, they reset the whitelist, and the mailing lists were dropped. By this point, MAW was contacting ADDR's support directly and reading them the riot act. They said there were still fine-tuning their spam filter.

Well, that was pretty much the final straw for me. As Paul Graham said:

False positives seem to me a different kind of error from false negatives. Filtering rate is a measure of performance. False positives I consider more like bugs.

So my email had become buggy, unreliable. I could no longer assume that any real mail to me would reach me. I was left wondering, did that other photographer never get back to me, or was his email just eaten by the server filter? My nice email account had suddenly lost a lot of its value, its assumed reliability.

So, I emailed ADDR.COM to request that this server-side filter be turned off for my account. They said that they could not turn it on or off on a per-account basis, but that they were working on it. I asked when they would finish with this upgrade in order to have that ability. They didn't know. So, they couldn't fix it today, and they didn't know when they could fix it. It was ultimatum time. I told them that if they couldn't give me a schedule for the fix within one week, then they didn't deserve my business. I waited one week and never got a response.

That weekend, I moved the first of my domains there, the experimental one. In response to my account cancellation, I received an email (my first since the ultimatum) which read:

It is extremely disheartening to learn that you would like to close your account with us. If it is due to any fault of ours, kindly accept my apologies. I would be grateful if you would let me know if anything needs immediate attention on our part. If there is something that needs to be altered to make you reconsider your decision, kindly intimate me about it. I assure you that it shall be my top most priority to set it right.

Oh yeah, I'll intimate them about it. Sufficed it to say, his offer of assistance was far too little, far too late. I simply referred him to the support ticket number and explained that I would not be reconsidering.

I should point out that IDCSoft does offer server-side spam filtering, but it's entirely configurable by the customer on a site-by-site and even mailbox-by-mailbox basis.

One last bit of oddities. I did some digging through the usage reports for this site, seeing what statistics IDCSoft would provide. These are all from September:

I still have one more domain to move, and that will be the trickiest since it has a message board, and I'd like to move the messages, too. I'll try transplanting it to the experimental one first as a test. Wish me luck.

Technology by Dan | Permalink | Comments (0)

September 02, 2004

Internet poles?

Philadelphia is consider making the entire city an internet hotspot:

The ambitious plan, now in the works, would involve placing hundreds, or maybe thousands of small transmitters around the city - probably atop lampposts. Each would be capable of communicating with the wireless networking cards that now come standard with many computers.

So, does that mean we're going to have guys who say that whenever they walk past a lamppost, the internet goes down?

Technology by Dan | Permalink | Comments (0)

August 16, 2004

Is this secure?

For work I frequently use a virtual private network. The login authentification is comprised of a short PIN number followed by a 6-digit time-based key. That key is provided by a small hardware token and is only valid for a minute or two. At the other end, my combined number is authenticated with a combination of my login ID (pointing into some data table) and the current time. It is effectively a single-use password and is considered to be very secure. I even travel with the token, but I keep the PIN prefix in my head.

Now, here's a guy who doesn't want to travel with the token, so he has a webcam pointing at it. Thus, his time-based key is public knowledge. However, this is still fairly secure since we don't know a) what network this guy is trying to log into, and b) what his prefix PIN is. Still, I don't think his IT department would be amused.

Technology by Dan | Permalink | Comments (0)

August 06, 2004

This law makes itself illegal

From LawMeme comes a little article on how the INDUCE Act is running afoul the Turing undecidability theorem, a proven fact in computer science.

I'm not quite sure what a court would do with a law that could be proven impossible to apply correctly.

Technology by Dan | Permalink | Comments (0) | TrackBack (1690)

August 05, 2004

Time scales...

Someone from the performance team at work passed this along:

Now 2GHz is a difficult thing to imagine for a human. Put simply that is 2 billion (Dr Evil pose) instructions per second at maximum throughput. So lets put this on our terms. Let's say one processor “clock cycle” is not 1/2,000,000,000 of a second but rather 1 second. On that scale accessing the nearby L1 on-chip cache takes 6 seconds, the off chip (L2) cache 2-3 minutes, and accessing main memory takes 3-4 weeks. Accessing the disk (just one disk access) by comparison takes a whopping 1 year on this timescale.

Technology by Dan | Permalink | Comments (1)

An anti-spam tactic...crossing the ethical line?

I read an article the other day about a local, and wealthy, spammer. He gets paid about $7 for every lead he can hand off about refinancing a mortgage or what-not. Then I began to wonder if the tactics of spam could be used to poison that revenue stream, but it seems that's probably crossing the line into unethical behavior.

Then I decided that just talking about it is probably still on this side of the line. I'll let someone else imperil his soul crossing the line. Call me an enabler. ;)

Every spam that can actually bring someone some profit has a link going back to someone that will kindly take your information or your money. I'm thinking in particular about the ones that take your information for a followup. These are the ones that banks and other traditional businesses pay spammers for, allowing them to push the ethical problem of spam off to the middleman. They can then take those hot leads and followup with a high probability of a sale. Hot leads like that are gold. But what if that information starting turning bad? What if they suddenly started getting a hundred times as many leads, but 99% of them were bogus?

It's the same quandary we face with our email boxes every day, a signal to noise ratio of 1:99. Except while our cost is just annoyance and wear and tear on our delete keys, the cost to these banks and other buyers of leads would be the time and effort of a followup call that dead-ended. The value of those hot leads would drop not just 100-to-1 but even lower. To get to that one hot lead, you'd have to grind through 99 cold calls. At that point, these spam-generated leads would be worthless.

Now, we could all just start following up on spam, filling in bogus information or perhaps even with valid information but for someone not interested in the product, but that's going to have a limited impact. Not many people would do it for very long, but what if it could be automated? That's the main tool of the spammers, so can it be turned against them? Could we write tools that followed spam links, looked for the right data-entry fields and plugged in the information? Such a tool could create semi-random data (Abraham Adams through Zachary Zane) or just pull it from a collection of real contacts such as those provided in telemarketing databases, available on CD-ROM for under $100.

Now, this would only be poisoning the leads that came through the spammers. They would already be tracked as such in order for the spammers to get paid, and as such, the buyers of the leads would be able to identify the source and cut it off. Spammers would have a record month, and then nothing except maybe a lawsuit or two.

Of course, not all spam falls into this category, so this is not the ultimate solution to the spam problem. But when I read about someone driving a Jaguar because my inbox is full of spam, I start thinking of ways to fuck with them.

Technology by Dan | Permalink | Comments (2)

July 24, 2004

Secure code through better development tools?

Here's an interesting article on producing better software through the use of better development tools, many of which are still just theoretical. It's an interesting read if you're in the software game.

Technology by Dan | Permalink | Comments (1)

July 23, 2004

Future alternative to Lasik surgery

I've mentioned this to a few folks before, but I never had a link to post. While Lasik results vary from exceptional to horrific, the real risk to me is that it's irreversible. If they get it wrong, it doesn't seem like there's much they can do. That's one of the things that's kept me from doing it. Also, most doctors I talk to tell me that getting to 20/40 is about all they can really guarentee, and I'm holding out for a 20/20 (or better) solution. After all, I'm currently in extended-wear contacts, so I'm already close to the level I want.

A few years ago my opthamologist told me about a new technique where they lift the cornea and place a contact lens beneath it and then seal things back up. The beauty of this is that it's reversible and repeatable. If they get it wrong or if your prescription changes, they can replace or remove the lens. The company that's doing much of the research is STAAR Surgical Company. Currently the procedure is not available in the U.S., but it is working through the FDA approval process. Like many Lasik patients though, I'll be holding out to be some doctor's 10,000th patient. No offense to the first 9,999, but my work and most of my hobbies depend heavily on my eyesight.

Then, holding out near the horizon is the possibility that a myopia/hyperopia combo lens could be implanted for those of us older folks whose arms aren't long enough any more.

Technology by Dan | Permalink | Comments (0)