Dr. Jerry Pournelle

Email Me


Why not subscribe now?

Chaos Manor Subscribe Now



Useful Link(s)...

JerryPournelle.com


Hosting by
  Bluehost


Powered by Apache

Computing At Chaos Manor

Mailbag for November 6, 2006
Jerry Pournelle jerryp@jerrypournelle.com
www.jerrypournelle.com
Copyright 2006 Jerry E. Pournelle, Ph.D.

November 6, 2006

I have been using Firefox as my main web browser for several months. You need to be careful which extensions to employ; some work differently depending on the hardware or perhaps other software. I have two machines running kept-up-to-date Windows XP, and on one of them the combination of Tab Mix Plus, Session Manager, and Crash Recovery works fine; on the other there can be problems with session recovery with that combination after an irregular shut down. If I figure out why those differ I'll tell you.

Note that I am using Firefox 1.5.0.7, and I'm in no hurry to update to Version 2, but I sure like it when others do and report it to me. Note that I haven't tried IE 7 yet either, but I will as part of doing things so you don't have to.

Subject: Firefox 2 vs. Internet Explorer 7

For a quick change of pace... Here's my initial take on the two new web browsers: The much talked about Internet Explorer 7, vs the shiny new Firefox 2.0.

Firefox 2 has been massively turbocharged; it's very fast. And Firefox 2 is still more secure than IE7; Ad-Aware scans show more spyware on the system after IE7 usage than after FF2 usage. You can easily run the check yourself, and I recommend you do so.

Both browsers now offer some degree of protection against phishing scams. The best protection is still a human brain, and I recommend its use.

FF2 maintains the same user interface, so there's no learning curve. With IE7, you need to learn to use the browser all over again. And hiding 95% of the commonly used controls in IE7 is just stupid. Was Microsoft trying to design a browser, or a puzzle?

You can add bookmarks to a toolbar in FF2, which speeds surfing to commonly viewed sites. IE7 doesn't have this.

FF2 is extensible, IE7 is not.

FF2 is written by gifted volunteers that have the public good uppermost in their minds. IE7 is written by Microsoft. Consider the implications.

However, the zoom feature of IE7 works better than the zoom feature of FF2. And it supports Activex, which is occasionally important (and often a huge security risk). For my purposes, Firefox is the browser of choice. Thats the news to this point...

Charlie Worton, MCSE, A+'

Thanks, Charlie. I'll join you on the frontier pretty soon...


Now a horror story with a happy ending. Clearly the first lesson here is Don't Do That! But it's well to know you can be made whole. Maybe.

Subject: Browser hijackers and pests

Hello Dr. Pournelle,

Here's a little bit of info that might be useful to some of your readers. I was browsing some of the darker regions of the web the other day and I tried to run a program I had downloaded. Annoyingly enough, it contained several Trojan Horse programs that immediately installed a system tray icon that told me I had viruses, hackers, etc. Yes, go ahead and say it, I was stupid for running the program. It looked like a legitimate link to a Microsoft plug in, but was NOT.

Anyway, I was stuck with some very annoying (and disturbing) pop-ups and the system tray nag telling me that I had viruses. If you right-clicked on the nag, it prompted you to buy a "virus removal" program. Well, I ran McAfee Anti-virus, Ad-Aware and Spybot Search and Destroy. Much to my chagrin, nothing killed the pop-ups or the nag-ware in the system tray. So much for conventional wisdom...

Anyway, at that point I got mad. I turned off System Restore to keep any unwanted pests from regenerating, then got to work. I found the "software plug in" I had installed, but attempting to delete the files while running XP in "normal" mode only gets you a steadfast refusal to delete the files from Windows XP.

I rebooted in Safe Mode and deleted the entire directory, files first and then the directory itself. Then I went into Regedit and deleted all references and entries associated with the file names found in the offending directory. That killed the pop-ups, but I still had the nag-ware telling me to buy the unscrupulous anti-virus program running in the system tray. So how do I find the offending files? Easily, really. Run an advanced search using Windows Explorer searching for all *.dll and *.exe files modified from the date of the infection to the current date (I had let a couple of days elapse due to other things I was doing). Copy down any file names meeting the criteria. Re-boot into Safe Mode, command line only using the F8 key during the boot sequence. Test the files by copying them to a bogus file name (I used zombie.txt), then delete the .dll files that met the search criteria one by one.

In my case it was easy; only the offending file had met the date criteria. I supect that will be the case in most occurrences like this. Reboot back into XP normally. Viola! No more nag-ware! Then delete all references in the registry to the offending file and delete the file that you had backed it up to. Obviously, if you have more than one file show up on your search, you'll need to test them one at a time. But it works, and it works when your anti-virus and adware removal programs have failed you. I hope this can save some of your readers from performing a total system restore when they otherwise wouldn't need to do so...

Very Respectfully,
Paul Freed

I suppose that might work, but I sure wouldn't trust that machine with any important data. When you get one of those persistent root kit infections, you are best off nuking from orbit, scrub to bare metal, and start over. It's inconvenient, particularly installing applications when you have mislaid the installation disks, but that's a lot safer.

The real moral of this story is to make good restoration point files using Norton Ghost, Norton Save and Restore (those are the ones I have used with satisfaction) or a similar program you have reason to trust. Then if you get an infection, you boot with the restoration CD or Floppy, reformat the hard drive, and then install the recovery file. That puts you back where you were before your adventure. And that way you have some certainty that the beasties are dead.


And now another horror story with a satisfactory ending:

Subject: Moving home - ISP nightmare & a useful trick

After 20 years I am moving home to "retire" my mortgage - I won't bore you with the horrors of dealing with incompetent lawyers but one aspect of this story may raise a wry smile & also pass on a useful trick.

I use an 8 meg ADSL connection provided by an ISP set up under the LLU (Local loop unbundling) rules that enable the "last mile" of copper previous owned by the British Telcom monopoly to be taken over and new equipment installed in the local exchange - so far so good 18 months of high speed access & telcoms. I use a Linksys ADSL2 modem feeding a Vonage (Linksys) VOIP enabled router to provide an additional business line and the whole lot accessible via a WPA protected WiFi network - perfect!

I contacted my ISP/telco to request a move of the service to my new house a mile away - well they replied informing me that they had instigated a cancellation of the service from the planned date of my move. Now for the problems, they were unable to transfer my telephone number to my new address, they could only take over the existing BT line at the new address - no good to me especially as they could only provide 2 meg on that line (1.5 miles from exchange V. 1 mile current) but if I requested a new line from them then an up to 8 meg service was available.

OK I'm not that attached to my 'phone number and would rather have the high speed link so I took their advice to cancel existing contract and apply for a newline at my new home. Now here's the thing - they can only take payment for the new connection via credit card and, you've guessed it, as they are registered to my current address they will not validate for a new order at my new address. I can hardly change all my bank & CC details ahead of actually legally becoming the owner of the new home so I cannot order the service until I move in so no 'phone or broadband for a month - Great!

The solution?

I have requested my existing 'phone number is transferred back to BT who will then be able to transfer it and I will have to order a new service from my ISP after I move (If I get a ISP service from BT then only 2 meg & minimum 1 year contract) in so although no broadband for a month at least I should keep my number & get an8 meg link from my existing ISP (they will refund the connection of a new line if order within 3 months.

The trick?

If have a Fujitsu Siemens ST101 tablet pc with a 3G umits roaming broadband card supplied by Vodafone - just upgraded from .4 to 1.8 meg, (I have a 250 meg monthly data allowance) I have managed to set this up using Windows ICS to replace the ADSL Modem & feed via it's Ethernet port straight into the router and voila! All is well, including VOIP line, if a little slower. Now I have proved this works I am ready to set this up as a fall back until my new connection is functional at my new home. Hope this useful to you or other readers.

Regards Andy Gibbs

I actually hope no one has to find it useful! But thanks!


Roberta recently was victim of what amounted to a denial of service attack, although it probably wasn't intended that way. As a result she got up to 10,000 "returned" emails a day; someone had faked return addresses at her domain as the return addresses to a lot of spam.

That story is told in the column, and has a happy ending, but while it was happening we got this letter which may be of interest to readers.

Subject: 10,000 returned emails

Jerry,

As a long time reader (Byte ~1981) I have pretty much always been getting good advice from you. Perhaps I can reciprocate.

With regard to the 10,000 returned emails. A couple of years ago, my wife's domain (also Mindspring/Earthlink based hosting) experienced a very similar problem. Besides the overflowing rejected email I was concerned that her domain would get placed on a black list and legitimate email from the domain would be blocked.

I decided to try to follow the money....

The spammer is advertising something - there is a link in those emails to some web site somewhere, even if it is only an IP address. I found out who registered the domain, and under what top domain it was located. Here are a couple of the registrars that I searched back then:

http://www.arin.net/whois/index.html
http://www.ripe.net/ripencc/pub-services/db/whois/whois.html
http://www.apnic.net/apnic-bin/whois2.pl
http://lacnic.net/cgi-bin/lacnic/whois

(I don't know how up-to-date these links are.)

I then sent a succinct email to the registrar and the domain reporting abuse, and asked the registrar to suspend or discontinue their domain.

I got a response from the registrar (it was registered in the Asian Pacific section) and the deluge did stop.

Good luck.

PS. I used to be bothered by your injection of political fare into your column but now I enjoy reading it, even if I tend to disagree with it. I have slowly discovered that you are not entirely dogmatic :-).

I'm fairly impressed with your pieces on Net Neutrality. I particularly enjoyed your inclusion of other points of view. I do agree that IF the market can solve the problem then that is the best way to do it.

There was a nice PBS show on the subject done by Bill Moyers "Moyers on America" called "The Net at Risk"

http://www.pbs.org/moyers/moyersonamerica/net/index.html

It has some interesting historical perspective on regulation of "common carriers" which I didn't know about.

There's much more but I digress.

Chris
--
G. Chris Boynton


Last week's column on net neutrality generated a fair amount of mail. Here's a sample.

Subject: "getting your Google access at 2400 Baud."

Throttling a popular site like Google to begin with is pretty farfetched. Suggesting that it would be throttling it to 2.4kbs implies to me that we are not dealing with a rational debate.

But let's assume a more likely worse-case scenario. Let's assume a 56kbs limit/channel for Google requests. A typical response page is about 24kB, not including the logo. So you are looking at an approximate 3.5 second download time after a query. It must be wonderful to live in a world where a 3.5 second page download time is considered something to raise arms about.

Gene Horr

Let's just assume John Dvorak was exaggerating for effect...

And Eric's comments on net neutrality came too late to incorporate in the column:

As it stands now, video delivery is pretty awful. GooTube stuff looks awful much of the time because they squeeze the hell out of material that didn't look very good to start.

Video delivery over IP is expected to be big money and people want it to perform the way their TV service does. (As it is, digital cable systems have added a lot of channel changing delay compared to the analog days.) One of the big drivers for the telcos to pull fiber to our homes is to compete with the cable Multiple Service Operators for the TV business. Electronic programming guides are all well and good but people want to roam around and see if something appeals. This means trying to match the performance that cable delivers now. Give people that and after a while they'll look at what we have now as the bad old days.

Video has always followed where audio led. Remember when audio streaming over the internet using dial-up was a remarkable thing? It was a big deal just a decade ago that we could pull up the audio stream of our choice. The quality was lousy but it kind of worked, sometimes. Nowadays, any decent broadband connection lets us choose high quality audio streams alongside other activity on the same connection. Plenty of people listen to online radio feeds while surfing other sites, downloading files, etc., rarely pausing to appreciate how amazing this would have been just a few years ago. Bandwidth made most of the difference but there were infrastructure needs beyond that, too.

People want interactive video with good real-time throughput. Webcams have been a staple in the PC world for a long time but now they're getting popular on game consoles with integration into their online services and games. Video conferencing will probably never be as popular as voice since no knows how awful you look by voice alone but the easier it gets the more people will expect it as a standard item and complain when it doesn't perform well.

The quality level requirements are only going to increase. Market penetration for HDTV is only just now getting up to snuff. Today, you can go out and buy HD disc players for retail content, video game systems designed for HD displays, and cable/satellite services with a decent selection of channels. There remains a lot of maturation ahead but for the average consumer HDTV is far more of a real product than it was a year ago. Video delivered over the net will need to keep up with the raised expectations HDTV will instill in consumers. Bandwidth is the biggest factor but you still need to use that bandwidth properly to create a competitive service.

Eric Pobirs


Subject: broadband competition

Re Eric Pobirs: Wi-Fi et al won't provide any competition because they all feed through a telco or cable, which has to be paid.

Unless you're in a Megalopolis, you have no competition in telco or cable. Read up on the reasons for Federal regulation of "natural monopolies" in the early 1900s. While there are many arguments being postulated as to the nonexistence of such monoliths, is your electric bill less in constant dollars than in 1980? (Yes, constant dollars is another myth; i.e., the depletion of the Atlantic cod fisheries and the artichoke plantings in CA make those parts of the market basket inconsistent.)

Why the consolidation in the three broadband industries? Could it be that the cost of serving all the little branches of the tree is too much? Could the New Paradigm be wrong? Should *BGAN* [Inmarsat's next-generation satellite broadband] be priced the same as VOIP?

Donald Miller

Note that there is already competition among satellite Internet providers...


And finally, a comment on the book of the month:

Subject: The Trouble with Physics

Jerry,

Your mention of this book in this week's column reminded me of the so far futile attempts to detect gravity waves. There is also an elaborate satellite based system under construction to try and detect gravity waves. I got to thinking, why is it we know so little about gravity and why haven't we been able to detect gravity waves?

Well, we live in 3 space plus T and haven't been able to detect any of those other dimensions that the string theorists use to make their equations work. (Of course there hasn't, to my knowledge, been any significant attempt to detect them.) What if gravity waves actually travel in those dimensions?

The best demonstration that these dimensions actually exist is entanglement. Entangle a pair of atoms and insert a photon into one and a photon comes out of the other instantaneously regardless of the distance between the pairs. I would say that entanglement connects the paired atoms via one or more of the extra dimensions so that they remain directly connect to each other regardless of their location in our 3 plus T space. In other words, it is possible for any point in the universe to be next to any other point in the universe.

Why haven't we been successful in detecting any transmissions in our SETI experiments? We are looking for RF emissions. RF is probably the crudest form of communication to an advanced intelligent race. It limits communication to the speed of light, is subject to interference, is not secure and RF emissions may interfere with brain activity or possibly cause cancer. It seems to me that any advanced race would make use of the dimension used by entanglement for communication. It is instantaneous regardless of distance, secure and produces no emissions that may be harmful to health!

There is a limitation to communication via entanglement. In order to establish communications you must first get to the distant point via some physical means of transportation.

The argument against entanglement as a means of communication is Heisenberg's. If you try to access the information in one half of the entangled pair, i.e. determine the spin of a photon, the entanglement is broken. However, you are merely capturing the emitted photon from the entangled atoms and not accessing any information in them. I think that it should work. If it doesn't then the probability of interstellar civilizations is rather small.

Bob Holmes

I thought Bob Forward at Hughes did detect gravity waves, but I will cheerfully admit I haven't been following this lately. When I was science editor of Galaxy and had to do a general science column every month, I was more careful. I often miss those days, and Baen Books is trying to get me to revive my general science essays on a quarterly basis. The problem is time, because I sure liked doing them. I also liked having a good excuse to keep up with the latest in science.

Some of my Galaxy columns were collected into a book published as A Step Farther Out. The rights long ago reverted, and I have just about got that together for publication as an ebook; they are, alas, still fairly current despite their age. And I have a second volume called Another Step Farther Out compiled from later columns; with luck both should be in electronic publication before the end of the year.