I don't like how the article says that only Windows software is vulerable.
Windows is just the most popular target for viruses. Eventually, as other platforms get popular (eg: Mac, Linux), hackers might take the time to port viruses to take advantage of commercial software across platforms. One example is Smile.d. As long as there are programs that let users run arbitrary code without screening it first, we'll still get viruses to take advantage of that vulnerability. Virus detection software is a good stopgap measure to help (ie: not completely) prevent idiots from hurting themselves through carelessess or lack of education.
As for the images being infected with viruses. I can imagine how someone in the future might develop an image format (maybe audio or video stream today) that allows one to embed commands as a convenience. I can also imagine how someone could take advantage of buffer-overflow in a poorly-written image viewer. I'm not saying someone has done it or that I know of any image viewers or media players out there that are vulnerable, but saying it's "not possible" is only right until proven otherwise.
% dig za soa za. IN SOA rain.psg.com. hostmaster.psg.com.
Go Randy!
BTW: In cases where a national government and an interested party are at odds over the control of a TLD, the government usually wins in the end. There's a saying, though, about ownership that goes, "Posession is 9/10 of the law." That's what keeps ".za" running right now. At some point, the SA government can/will exert its influence to either 1) get ICANN to change the TLD NS records or 2) outlaw the use of unofficial ".za" servers for any SA ISP. Perhaps the best scenario will have Mike and the SA government reach a transition compromise. If not, it'll be interesting to watch in a Death-Star-finds-rebel-base kind of way.
We are taking all of the addresses and domain names back later this week. Please stop using them now. We give you permission to use any address starting with the octet of 10. and register names under the DOT top-level domain. To register, please visit our subscription page before 23:59 GMT Friday June 7th.
I was on an airline descending into San Diego during the eclipse. The ground had an eerie feeling from above, though the sun was bright as usual. Using the "pinpoint" trick, I could see the eclipsed-shaped sun in its shadow.
When we got on the ground a parking agent was kind enough to lend me their translucent viewer. Sure enough, 40% of the sun was "eaten" by the moon. It's always fun to look at.
Being close, but not close enough to an eclipse at sunset, I wonder if anyone in Phoenix, Tuscon, or Mexico got any cool sunset views or photos during their sunset. Tonight I saw the full sun set an hour after the eclipse and wondered how cool it would be to see an eclipsed sun setting. Anyone? Put your pics online and give us a URL for some surefire moderator points.
There's a term used by many in the storage industry to get you investors excited about storage companies. It's called the "explosion of data". Because we are able to record so much data with so much detail, we do. Like the pack rats we are, we think it might be useful to us in the future. That's what the storage vendors are hoping for.
The following may be absurd, but (in a manner similar to Carl Sagan's Cosmos series) it may help enlighten us as to how much detail we don't see and don't collect about a particular event.
Instead of the entire universe, let's take a look at a World Cup soccer/football game.
The best compressed description of the game might be, "Italy Defeats Equador, 2-0".
Add a little more detail, and you might have an accounting of highlighted events from the game. For example, here's one article.
They say a picture is worth a thousand words. Here's a picture. During the match, spectators probably took several thousand pictures.
After the match, somoene probably went with a tape recorder and got quotes from each of the players. The audio is alot of information, but compressed, heres a text summary.
During the match, there were probably 10-30 professional video recorders sponsored by televsion stations in Italy and/or Ecuador that would provide a live satelite broadcast of the match for fans back in their home countries. The fans only saw one view of the data, the finished product. Each camera probably had its output recorded on tape. Out of the millions of viewers, a thousand of them may have recorded the match on video casssette. Inside the stadium, hundreds of fans probably used their own recorders. The amount of storage, accurately reproduced might total terabytes of data, yet this is only a fraction of the number of possible views that could have been recorded.
The player of the game, Italy's Christian Vieri, must have been a crucial part of the win. What's his life's story? What events made him what he is today? Could we video his entire life (ala "The Truman Show"?). Can we understand all of his thoughts just during the game and why those thoughts occurred? What were his vital signs each minute of the match? How did his movements or actions affect each and every other player, physical object, and every fan who saw him play that day? What were the contents of his upper intestine? From which oil wells did the petroleum needed to make his shoes come from? How much energy did he expend? When he shot his goal(s), and all of those Italians in bars cheered, what was the effect of those cheers on the microclimates in each of the homes and bars where he was viewed? If one of the players he bumped into developed a bruise on his thigh, how many blood cells left circulation to stagnate in that area? What was the percentage of hemoglobin in those lood cells? What signals were sent to the brain and in which order were they recieved by which receptors to help trigger the player's lymph system clean out and heal that area? If we took one of the millions of hairs from his head and analyzed it, would we be able to find that he smoked marajuana a month ago? What is the data from all prior events over all time that was relevant to creating Mr. Vieri as he is today?
What about the other thousands of people in the stadium? What are their stories? Where were their clothes manufactured and with which materials in which locations? Who developed the film on their cameras and what are their life stories? As they breathed, where did all of the carbon dioxide atoms that they expelled end up a few seconds later, a day later, a year later?
What was the placement of each atom of the concrete and steel to create the stadium? What was the position of each blade of grass and molecular composition of each blade over each second of the game? Show me the path each electron took and its final position at the end of the match.
We cannot come close to understanding, though, the amount of data necessary to "record" that event. It is only through selective compression, what our senses tell us, that we develop our view of that event. For some, like Mr. Vieri, he may remember what he felt and experienced during and after that event. A fan in Italy might remember what they saw, and might even have a tape or picture that shows one view of the event. A sports writer in Equador might only remember that Italy beat Ecuador 2-0. The average person on planet Earth will have no knowledge or recollection of the event, and frankly, won't care because life is too short.
Good analysis of events is compression. Our memory is compression of our experiences. With good compression, we won't have to record everything, and therefore avoid the "explosion of data" as best we can. As we collect data, we need to consider its importance to us and discard anything not relevant.
For detail we do care about (eg: data needed to compute Earth's weather), we might try to build large data repositories and build expensive computers to process that data, but most of the Universe's data is best left unknown to us because it's not important to us (yet).
- ez
PS: You gotta hand it to the folks at Google for attempting to collect and store so much data from the Internet.
Am I the only old fart left that uses MH from the command line?
If I recieved Slashdot Poll postings as e-mail messages, I might use the following to find recent common whining about lame slashdot poll choices (and faster too!):
cd `mhpath +slashdot` pick -subject CowboyNeal | xargs egrep -i 'this poll sucks' | sed -e 's/:.*//' | uniq | xargs show
Go to the link above, or look for MH or NMH in rpmfind.net or your local ports tree.
"Artists, in particular, are hurt because under most recording contracts, promotional costs come out of their royalties", said Michael Bracy of the Future of Music Coalition.
Why? I thought it was the record company's job to promote music. Shouldn't that come out of their percentage? If it's not worth promoting enough to make a profit, they probably shouldn't have signed a contract with the artist.
What happens when the record label decides to do a couple $100,000 promotional parties^H^H^H^H^H campaigns on relatively unpopulated tropical islands? "Er, sorry, but your promotional expenses exceed your royalties. You owe us money."
The small (size) is of course very important, but it is a little bit overhyped. It is really the performance we are after," said Phaedon Avouris, manager of nanoscience and nanotechnology for IBM Research.
Who says size doesn't matter? As long as the electrons are satisfied!
It really pisses me off when some dumb son of a bitch thinks they have the right to directly send an email from their dialup account.
It's possible for CmdrTaco to insist that people he talks with on a regular basis recieves mail from him. He can also configure a mail server at Andover (or any of his friends' servers) to always accept mail from him and use it to relay his mail (hopefully spammers don't find out what that network is and do the same).
It's also possible for people to just stop using e-mail to communicate. Got a project people need to work on? Setup a private web-based mail system. Everyone logs in to a common website and sends messages internally related to a project. Even better, use a threaded one-to-many conferencing system like Slash.
If you don't like spam, stop using e-mail. My main mailbox sits at the end of probably the most restrictive filtering system on the net, and I still get spam and viruses sent to me.
Regarding MTA registration
Frankly I think there should be a mandatory MTA registration process like there is for DNS servers. You must register your MTA before you can mail. This would require valid abuse contact information and things of that nature.
Instead of mandatory MTA registration for all mail servers (Hey! Verisign pioneered registered certificates for Web servers, hmm...) perhaps it's better to start other lists of servers that get an explicit "allow" like 1) "These are the mail servers of my friends", or 2) "These are servers on the Better Mail Server Bureau list becausee they adhere to policies that help stamp out spam and are accountable."
It all comes down to "who do you trust"? Back when the Internet started, everyone was like-minded and trusted others not to screw up the network. Now the circle of trust is much much smaller.
Another good place to filter spam is at the mailbox level. My mail server hypothetically accepts e-mail from everyone. Messages from people in my address book go straight to my Inbox. Mail from everyone else goes into my junk inbox. I have mechanisms that easily allow me to add addresses to my address book. Spam becomes less annoying becasue I don't have to deal with it while reading the mail I care about.
Email is antiquated
Instead of putting people on mailing lists, subscribe them to bulletin boards. Mail clients get replaced with web browsers. Sending mail becomes posting on your own bulletin board and giving me a URL (/w messenger) to read it.
The architects of Usenet2 had a great idea in that people posted messages on their servers and interested people would have to fetch the articles to read them. If you have popular content, the effect could be similar to SlashDotting. Smart people would enable caching for larger reader communities. Instead of broadcasting trash everywhere, people would pick up the content they want. The economics of spam would dissappear. Applied to email, spammers would merely send me headers. It would be up to me whether I wanted to pick up their "content". I might filter headers similar to how I limit my Instant Messager alerts to friends.
Aside
Can't we all just get along? Ok, perhaps we can't. We'll just go off and limit our interactions to the people we like (or can at least tolerate). That's what people have done in the past, and that's what people will continue to do. Laziness in the implementation of Internet e-mail today just makes it harder for us to keep out the people we don't want to interact with. This "everyone in the world can send me e-mail" concept is much overrated. It was novel in the 80's and 90's. People who can't e-mail me now (I live mostly behind a MAPS firewall) can pick up the phone and call me or reach me through a friend or find me via a spam-friendly free e-mail account.
There's no doubt that the movie is having a successful weekend, but how successful was it?
Is the John Harman a prophet? or is he just a part of the hype machine for Sony? He already seems to have wrapped up the weekend in past tense before it's even over.
Spider-Man opened to $114 million on 3,615 screens
At least the Yahoo article quoted sources:
... according to studio estimates issued on Sunday.
Let's take it for what it's worth - propaganda. The goal is to get the people out there thinking, "Gosh, this movie is so popular. Maybe I should go out tonight and see it."
The weekend is not over. Sony could hypothetically be ready to announce next weekend's box office results on Thursday this week. We'll all forget about Spider Man the following weekend when it's 15 minutes of hype^H^H^H^H fame are over when next Star Wars prequel is released.
What movie company was beind movies like "The Animal" that garnered rave reviews from fictional critics?
In "A Beautibul Mind", the lead character imagines a different reality where he's saving our country. All of us Slashdotters can just as easily collectively imagine that LotR won best film, actors, and supporting actors last night. Hopefully, no one out there will try to give us shock therapy to correct our illness.
Blue Mountain's mistake is thinking their product is virtual greeting cards. It's not.
Unlike before, the sending of cards are now a significant part of the product.
The original business model for BMA actually was hard e-commerce. The free e-cards were the loss leader to generate traffic. The bet was that a small percentage of the millions of users would attach high-margin gifts to the cards. The owner of BMA started an online flower company, worked with a chocolate company, and started a gift certificate service. BMA was going to go public based on revenue derived from the hard e-commerce sites. For its huge success, the site was also cheap to run. It had stable paper card revenue to support it (much like Hallmark an American Greetings funded their sites).
Several companies looked at BMA, and Excite@Home won the opportunity to buy it, offering $800m of funny-money and $350m cash based on their perceived value of BMA being able to boost their Internet reach and indirectly boost Excite's advertising revenues. Soon after Excite bought BMA, Excite tried pimping the site out to ordinary advertisers, but the floor suddenly fell out of the Internet advertising business. While BMA wasn't losing as much as competitors, they weren't necessarily profitable either. Yahoo, DoubleClick, Excite and others were hit incredibly hard in the advertising fallout. Egreetings went bankrupt, and Hallmark and American Greetings were leaking money at a pace of $25m-$35m per quarter. It was becoming obvious that the $1b investment wasn't ever going to materialize for Excite. The founders of BMA, though, were quite a success story.
With Excite on its way to backruptcy, American Greetings snatched up BMA for pennies on the dollar ($35m?) and then instituted a pay-to-send model. Less traffic from non-paying customers reduces operating expenses (especially on peak holidays), and the cards the users send are now linked to real money. The cards are now the product to their paying users. There are still some free cards available, though, and one can still see ads and attach gifts on the free cards.
Is BMA a failure? No. It's been the best-run and most efficient greeting card site. It's high-flying glory days are over, though. The flaw is in the gamble of those who paid an absurd amount of money for something not proven to be able to have a return on its investment.
The users who pay BMA yearly are the ones who send e-cards frequently. The last-minute Valentine's Day lusers will go elsewhere and help another free greeting card site lose their money faster.
How many people using Slashdot would pay for it if they had to? I would. Along with Google and MyYahoo, SlashDot is one of the few sites I enjoy regularly. Would I pay for cnn.com or pogo.com? No, becasue I don't use them frequently. Others might.
Just like there's basic cable and extended cable, Internet content sites can be lumped together to colectively provide service to paying customers. Zillions pay $9/month or more to AOL. They may not appreciate all of the services, but their money helps fund them, even the greeting cards.;^)
... is that there are so many of them. We've all heard this before.
Sure, IETF can create a standard, but it won't stop AOL or Microsoft from doing what they think is best for their user groups. Hackers will have their own AES-encrpted IRC VPNs. The smarter implementations will figure out how to use caching and flow control to reduce bandwidth demands. The ones with more money will have servers to support and scale these networks. The cheaper ones will rely upon more peer-to-peer.
Pssst, hey buddy... come here a little closer so I can tell you all quiet-like what you're gonna do.
First, find a credible University with poor computer security in their administration department, and then hack yourself in to look like you've been a transfer student with good grades. You can then take a final semester of the interesting classes you need to complete your degree.
Yeah that's the plan. Real easy and all. Just take after me - I'm a bonafide PhD!
Cat 5 can easily handle 100Mbps traffic and phone traffic. Remember when Cat3 was the new thing compared to old Ma Bell twisted pair? That got stuck in alot of homes as phone wiring before 100baseT came out. Those people are now limited to 10Mbps or AnyPoint-like technologies. People who installed Cat5 are happy today. People who install Cat5 today will be happy today, but what about tomorrow?
There may be a day when some broadband provider drops off a >100Mbps cable to your curb for $50/month. There may be a day when home entertainment will take advantage of Gigabit Ethernet or Infiniband over twisted pair or whatever the next technology is. I recommend that you get spools of Avaya Gigabit now. The price difference is only about $50-$60 per spool over Cat5, and you're probably lumping this expense into a mortgage, so you're not likely to see the difference. Also, it assures you that you're putting in quality cabling. You have no idea what brand/quality of Cat5 cable your contractor is going to install.
Find yourself a good contractor. This contractor will have installed many homes before and might even have some active tract home contracts now that Cat5 is catching on. They'd not specialize in just Cat5, but also coax, telco, and security wiring. They would also make sure they install not just the cable, but the right jacks and wall plates at each location. I can't tell you how much time it wastes if you have to install your own jacks.
Have one set of twisted pair go mostly to a phone punch block (110 or 66, doesn't matter) and have the other be terminated (TIA-568A) into a TIA-568A 110-block with ethernet RJ-45 outlets.
Consider also instaling dual RG6 (coax cable) while you install your dual ethernets. Just like your ethernet an phone wiring, you'll want to centralize the place from where you want your lines run. You never know if/when your cable company will require two cables to get more channels. You never know if you or the next owner will become a satellite+cable channel junkie.
As for fiber, well, there aren't many proven applications for it, and it's more expensive, and there are fewer people out on the market who can do it right.
I think someone else mentioned using metal conduits around your wiring. I agree, though it might be pricey.
If your wiring is very important to you, and if you plan to spend more than 5 years in this house, the extra money you pay in your mortgage to get the job done right is well worth it. If you plan to flip the house after two years, just install two cat5 and one coax to each bedroom and each work/play area to be done with it.
Also, don't forget elsectrical outlets! It sucks to have all your ethernet bundles available in a space where you have no electrical outlets for your network gear and computers. Doh! The outlet near the central point of your wiring be better off with it's own 15-amp fuse.
-ez
PS: As an alternative, there's always wireless. Consider how/where you'd want to install wireless bridges. Consider the security implications, too.
I'm noticing >500ms ping times to a local router beyond my cable box - a symptom that I've seen happen before they disable modems. Upon calling Adelphia (Carlsbad, San Diego, CA), they said that they're sutting off today.
It should be coming back online on/around Dec 16th as a new local service.
So I went though all the domains, and noticed that only z.com seems to be a functioning website (redirects to Nissan's website.)
I was the first to register Z.COM. IANA once gave a directive that said, "all one-letter names shall be reserved to enable name-hashing at a later time". Working for a company that registered domain names on a daily basis, I thought, "If X.ORG can have a domain name, why can't I register Z.COM?" To my surprise, it worked! The following month, IANA gobbled up all the rest of the one-letter names.
A few years later, I started having people knock on my door monthly saying they'd buy or trade my domain. They didn't see much of a value to it, and neither did I. While I was a bit altruistic, I did have a price in mind where I'd do away with my domain. One day this guy offered me 50% more than that price, so I took it. It went toward a down payment on a house that later made me some real money.
The guy tried to make a simple Z.COM web portal out of it. Their gimmick was that all one had to do was hit "z" on their web browser address, and poof, there you were at Z.COM. The portal never gained momentum.
Other people bought it from him and tried again to make a portal out of it, but their gimmick was to give "lifetime" e-mail accounts if they visited the portal regularly. Again, another Z.COM portal failed, and those "lifetime" addresses disappeared with it.
The next purchaser was apparently IDEAlab. They never did anything with it and with their financial demise probably thought they should sell/dump it for whatever they could.
Enter Nissan. My guess is that they might release or re-release a "Z" car in the future.
I mildly regret selling the name away. I thought the purchaser would have done something better with it. I could give Nissan a web redirect as good as anyone else.
I was initially pissed off, too. If Jon Postel (the original maintainer) were still alive, I wonder if anyone could have laid a commercial hand on the US TLD. I once tried in the early 90's to get a non-RFC US zone registered and met heavy resistance from Jon. I gained new respect for his keeping the purity of that domain.
Interestingly enough, NeuStar is headed by a bunch of folk that include an IETF dignitary or two (eg. Bill Manning).
The following document details how they're going to implement everything. Section B-3 goes into what you can and can't register under ".us" as well as gives info about the trademark "sunrise" period. At least it looks like they did their homework. There won't be any "com.us" or "xxx.us", but "screw.us" and "slashdot.us" seem quite feasible.
While I don't agree with them in principle, it looks like it could have been worse.
Also, it's quick to implement. Any reasonable sysadmin can make a NAS server out of a machine and OS they already know. Installing and administering a NAS server is what the masses are trying to avoid when they buy an "appliance". In many cases, a NAS appliance is simply a server with a quick installation process and an easy to use configuration interface.
Here's what I think everyone uses...
Procom (started with BSD, FreeBSD?)
VA Linux (pre-implosion, based on Linux with ext3 & Mylex extensions)
Cobalt (Linux)
Sun (remember the Netra NFS server? Solaris)
Dell (Microsoft-powered NAS)
Compaq (Microsoft-powered NAS)
Maxtor (Microsoft-powered NAS)
Novell (yes, they have one, supposedly based on top of Netware-derived platform)
Netapp (custom OS, but suspect it was BSD-derived in early years)
EMC Clariion (VxWorks + Crosstor)
EMC Celerra (own RTOS + Linux control station)
MTI (older Crosstor on their own RTOS?)
Auspex (own RTOS[?] supported by Windows or SunOS control station)
AMI (FreeBSD)
... and countless others.
The underlying theme is "go with what you know". Each has their benefits and drawbacks. The companies have to support the products they create. What OS they know how to support probably influenced the underlying OS they used to implement their product.
AMI knows motherboards, BIOS, and RAID controllers. The software needed to make their hardware into a server was already available with FreeBSD and ported applications. Not that it's a trvial task, but all that they needed to add was a GUI, maybe a front panel, a simple installation process and some support for the product to keep their customers from having to "administer" a "server". Instead, the customers "configure" their "appliance".
The AMI products seem like good solutions for workgroups, but some caveats might apply:
- redundancy - seem to have little to no high availability features outside of the RAID controller
- heat - installing a few here and there are fine, but thinking you can easily put 320 drives in a 40U rack might not be prudent
- backup costs - Veritas NDMP slave licenses aren't cheap (but perhaps others are?)
- performance - there are higher-end products on the market for those that need more throughput or I/O operations per second
If we've learned anything fom the Internet, it's that people will always find alternate paths around censorship. Some new USENET groups will inevitably pop up to replace the need of the former groups. For example, alt.binaries.penthouse could find a new home in alt.religion.kibology.priestess. Other distribution formats (eg: cache pre-population of encrypted content) could eventually replace USENET binary distribution.
If Excite@Home had a spine, they might have phrased it like this...
Due to our lack of operating capital and the fact that we really don't have the staff and resources to the administer stupid content that some of our users like to download, we're discontinuing support for the following newsgroups:...etc...
You can go somewhere like Giganews if you really see value in it.
Were they smart and also had some balls, they'd stop supporting USENET, period. Instead, they'd give each customer of both @Home and GigaNews a $5/month credit to use GigaNews. In return, GigaNews could pay @Home $XX,000/month for better bandwidth to the @Home user base. @Home would additionally free up expensive storage/servers/staff resources and be done with any liability with USENET.
Here's a post I made a long time ago that tells you a little bit about how alt.* newsgroups work...
ALT FAQ
Disclaimers: I'm a former Excite@Home employee. For that, you can label this post as a Troll. The advice, though, about USENET outsourcing is applicable to many ISPs. I have no affiliation at all with GigaNews.
Personally, I like wooden mailboxes the best; you know, the ones that look like a little bird house?
Were I more cool, I'd put a little X.10 wireless sensor in it to let me know when it has been opened. I might link it to an e-mail/pager that says "You have mail!".
Just kidding. I've been using MH forever (late 80's). It's just so darn flexible for programmers to search their e-mail. I like my ability to search with Unix command line tools through zillions of messages I've sent/received to anyone. I've been using it since 1988. For MIME processing, mhn isn't that great, so I just bounce copies to Yahoo or Mail.com mailboxes so I can view them with my web browser du jour and then delete them. Instead of POP or IMAP, I suck my mailbox from a mail server using ssh and incorperate mail into MH on my local secure desktop.
Slashdot Mirror
I don't like how the article says that only Windows software is vulerable.
Windows is just the most popular target for viruses. Eventually, as other platforms get popular (eg: Mac, Linux), hackers might take the time to port viruses to take advantage of commercial software across platforms. One example is Smile.d. As long as there are programs that let users run arbitrary code without screening it first, we'll still get viruses to take advantage of that vulnerability. Virus detection software is a good stopgap measure to help (ie: not completely) prevent idiots from hurting themselves through carelessess or lack of education.
As for the images being infected with viruses. I can imagine how someone in the future might develop an image format (maybe audio or video stream today) that allows one to embed commands as a convenience. I can also imagine how someone could take advantage of buffer-overflow in a poorly-written image viewer. I'm not saying someone has done it or that I know of any image viewers or media players out there that are vulnerable, but saying it's "not possible" is only right until proven otherwise.
-ez
BTW: In cases where a national government and an interested party are at odds over the control of a TLD, the government usually wins in the end. There's a saying, though, about ownership that goes, "Posession is 9/10 of the law." That's what keeps ".za" running right now. At some point, the SA government can/will exert its influence to either 1) get ICANN to change the TLD NS records or 2) outlaw the use of unofficial ".za" servers for any SA ISP. Perhaps the best scenario will have Mike and the SA government reach a transition compromise. If not, it'll be interesting to watch in a Death-Star-finds-rebel-base kind of way.
-ez
(Former TLD zone admin for several countries)
We are taking all of the addresses and domain names back later this week. Please stop using them now. We give you permission to use any address starting with the octet of 10. and register names under the DOT top-level domain. To register, please visit our subscription page before 23:59 GMT Friday June 7th.
Sincerely,
hostmaster@nic.dot
I was on an airline descending into San Diego during the eclipse. The ground had an eerie feeling from above, though the sun was bright as usual. Using the "pinpoint" trick, I could see the eclipsed-shaped sun in its shadow.
When we got on the ground a parking agent was kind enough to lend me their translucent viewer. Sure enough, 40% of the sun was "eaten" by the moon. It's always fun to look at.
Being close, but not close enough to an eclipse at sunset, I wonder if anyone in Phoenix, Tuscon, or Mexico got any cool sunset views or photos during their sunset. Tonight I saw the full sun set an hour after the eclipse and wondered how cool it would be to see an eclipsed sun setting. Anyone? Put your pics online and give us a URL for some surefire moderator points.
-ez
The following may be absurd, but (in a manner similar to Carl Sagan's Cosmos series) it may help enlighten us as to how much detail we don't see and don't collect about a particular event.
Instead of the entire universe, let's take a look at a World Cup soccer/football game.
We cannot come close to understanding, though, the amount of data necessary to "record" that event. It is only through selective compression, what our senses tell us, that we develop our view of that event. For some, like Mr. Vieri, he may remember what he felt and experienced during and after that event. A fan in Italy might remember what they saw, and might even have a tape or picture that shows one view of the event. A sports writer in Equador might only remember that Italy beat Ecuador 2-0. The average person on planet Earth will have no knowledge or recollection of the event, and frankly, won't care because life is too short.
Good analysis of events is compression. Our memory is compression of our experiences. With good compression, we won't have to record everything, and therefore avoid the "explosion of data" as best we can. As we collect data, we need to consider its importance to us and discard anything not relevant.
For detail we do care about (eg: data needed to compute Earth's weather), we might try to build large data repositories and build expensive computers to process that data, but most of the Universe's data is best left unknown to us because it's not important to us (yet).
- ez
PS: You gotta hand it to the folks at Google for attempting to collect and store so much data from the Internet.
If I recieved Slashdot Poll postings as e-mail messages, I might use the following to find recent common whining about lame slashdot poll choices (and faster too!):
Go to the link above, or look for MH or NMH in rpmfind.net or your local ports tree.
-ez
I found this part to be confusing:
"Artists, in particular, are hurt because under most recording contracts, promotional costs come out of their royalties", said Michael Bracy of the Future of Music Coalition.
Why? I thought it was the record company's job to promote music. Shouldn't that come out of their percentage? If it's not worth promoting enough to make a profit, they probably shouldn't have signed a contract with the artist.
What happens when the record label decides to do a couple $100,000 promotional parties^H^H^H^H^H campaigns on relatively unpopulated tropical islands? "Er, sorry, but your promotional expenses exceed your royalties. You owe us money."
-ez
An interesting quote from the article...
The small (size) is of course very important, but it is a little bit overhyped. It is really the performance we are after," said Phaedon Avouris, manager of nanoscience and nanotechnology for IBM Research.
Who says size doesn't matter? As long as the electrons are satisfied!
-ez
In case people are looking for it, here's a link to a story describing the funeral along with a link to the video itself. It's pretty amusing.
Regarding CmdrTaco's "problem"
It really pisses me off when some dumb son of a bitch thinks they have the right to directly send an email from their dialup account.
It's possible for CmdrTaco to insist that people he talks with on a regular basis recieves mail from him. He can also configure a mail server at Andover (or any of his friends' servers) to always accept mail from him and use it to relay his mail (hopefully spammers don't find out what that network is and do the same).
It's also possible for people to just stop using e-mail to communicate. Got a project people need to work on? Setup a private web-based mail system. Everyone logs in to a common website and sends messages internally related to a project. Even better, use a threaded one-to-many conferencing system like Slash.
If you don't like spam, stop using e-mail. My main mailbox sits at the end of probably the most restrictive filtering system on the net, and I still get spam and viruses sent to me.
Regarding MTA registration
Frankly I think there should be a mandatory MTA registration process like there is for DNS servers. You must register your MTA before you can mail. This would require valid abuse contact information and things of that nature.
Instead of mandatory MTA registration for all mail servers (Hey! Verisign pioneered registered certificates for Web servers, hmm...) perhaps it's better to start other lists of servers that get an explicit "allow" like 1) "These are the mail servers of my friends", or 2) "These are servers on the Better Mail Server Bureau list becausee they adhere to policies that help stamp out spam and are accountable."
It all comes down to "who do you trust"? Back when the Internet started, everyone was like-minded and trusted others not to screw up the network. Now the circle of trust is much much smaller.
Another good place to filter spam is at the mailbox level. My mail server hypothetically accepts e-mail from everyone. Messages from people in my address book go straight to my Inbox. Mail from everyone else goes into my junk inbox. I have mechanisms that easily allow me to add addresses to my address book. Spam becomes less annoying becasue I don't have to deal with it while reading the mail I care about.
Email is antiquated
Instead of putting people on mailing lists, subscribe them to bulletin boards. Mail clients get replaced with web browsers. Sending mail becomes posting on your own bulletin board and giving me a URL (/w messenger) to read it.
The architects of Usenet2 had a great idea in that people posted messages on their servers and interested people would have to fetch the articles to read them. If you have popular content, the effect could be similar to SlashDotting. Smart people would enable caching for larger reader communities. Instead of broadcasting trash everywhere, people would pick up the content they want. The economics of spam would dissappear. Applied to email, spammers would merely send me headers. It would be up to me whether I wanted to pick up their "content". I might filter headers similar to how I limit my Instant Messager alerts to friends.
Aside
Can't we all just get along? Ok, perhaps we can't. We'll just go off and limit our interactions to the people we like (or can at least tolerate). That's what people have done in the past, and that's what people will continue to do. Laziness in the implementation of Internet e-mail today just makes it harder for us to keep out the people we don't want to interact with. This "everyone in the world can send me e-mail" concept is much overrated. It was novel in the 80's and 90's. People who can't e-mail me now (I live mostly behind a MAPS firewall) can pick up the phone and call me or reach me through a friend or find me via a spam-friendly free e-mail account.
-z
There's no doubt that the movie is having a successful weekend, but how successful was it?
... according to studio estimates issued on Sunday.
Is the John Harman a prophet? or is he just a part of the hype machine for Sony? He already seems to have wrapped up the weekend in past tense before it's even over.
Spider-Man opened to $114 million on 3,615 screens
At least the Yahoo article quoted sources:
Let's take it for what it's worth - propaganda. The goal is to get the people out there thinking, "Gosh, this movie is so popular. Maybe I should go out tonight and see it."
The weekend is not over. Sony could hypothetically be ready to announce next weekend's box office results on Thursday this week. We'll all forget about Spider Man the following weekend when it's 15 minutes of hype^H^H^H^H fame are over when next Star Wars prequel is released.
What movie company was beind movies like "The Animal" that garnered rave reviews from fictional critics?
In "A Beautibul Mind", the lead character imagines a different reality where he's saving our country. All of us Slashdotters can just as easily collectively imagine that LotR won best film, actors, and supporting actors last night. Hopefully, no one out there will try to give us shock therapy to correct our illness.
Unlike before, the sending of cards are now a significant part of the product.
The original business model for BMA actually was hard e-commerce. The free e-cards were the loss leader to generate traffic. The bet was that a small percentage of the millions of users would attach high-margin gifts to the cards. The owner of BMA started an online flower company, worked with a chocolate company, and started a gift certificate service. BMA was going to go public based on revenue derived from the hard e-commerce sites. For its huge success, the site was also cheap to run. It had stable paper card revenue to support it (much like Hallmark an American Greetings funded their sites).
Several companies looked at BMA, and Excite@Home won the opportunity to buy it, offering $800m of funny-money and $350m cash based on their perceived value of BMA being able to boost their Internet reach and indirectly boost Excite's advertising revenues. Soon after Excite bought BMA, Excite tried pimping the site out to ordinary advertisers, but the floor suddenly fell out of the Internet advertising business. While BMA wasn't losing as much as competitors, they weren't necessarily profitable either. Yahoo, DoubleClick, Excite and others were hit incredibly hard in the advertising fallout. Egreetings went bankrupt, and Hallmark and American Greetings were leaking money at a pace of $25m-$35m per quarter. It was becoming obvious that the $1b investment wasn't ever going to materialize for Excite. The founders of BMA, though, were quite a success story.
With Excite on its way to backruptcy, American Greetings snatched up BMA for pennies on the dollar ($35m?) and then instituted a pay-to-send model. Less traffic from non-paying customers reduces operating expenses (especially on peak holidays), and the cards the users send are now linked to real money. The cards are now the product to their paying users. There are still some free cards available, though, and one can still see ads and attach gifts on the free cards.
Is BMA a failure? No. It's been the best-run and most efficient greeting card site. It's high-flying glory days are over, though. The flaw is in the gamble of those who paid an absurd amount of money for something not proven to be able to have a return on its investment.
The users who pay BMA yearly are the ones who send e-cards frequently. The last-minute Valentine's Day lusers will go elsewhere and help another free greeting card site lose their money faster.
How many people using Slashdot would pay for it if they had to? I would. Along with Google and MyYahoo, SlashDot is one of the few sites I enjoy regularly. Would I pay for cnn.com or pogo.com? No, becasue I don't use them frequently. Others might.
Just like there's basic cable and extended cable, Internet content sites can be lumped together to colectively provide service to paying customers. Zillions pay $9/month or more to AOL. They may not appreciate all of the services, but their money helps fund them, even the greeting cards.
- ez
... is that there are so many of them. We've all heard this before.
Sure, IETF can create a standard, but it won't stop AOL or Microsoft from doing what they think is best for their user groups. Hackers will have their own AES-encrpted IRC VPNs. The smarter implementations will figure out how to use caching and flow control to reduce bandwidth demands. The ones with more money will have servers to support and scale these networks. The cheaper ones will rely upon more peer-to-peer.
Does anyone miss BITNET RELAY? I sure do.
- ez
Pssst, hey buddy... come here a little closer so I can tell you all quiet-like what you're gonna do.
First, find a credible University with poor computer security in their administration department, and then hack yourself in to look like you've been a transfer student with good grades. You can then take a final semester of the interesting classes you need to complete your degree.
Yeah that's the plan. Real easy and all. Just take after me - I'm a bonafide PhD!
Cat 5 can easily handle 100Mbps traffic and phone traffic. Remember when Cat3 was the new thing compared to old Ma Bell twisted pair? That got stuck in alot of homes as phone wiring before 100baseT came out. Those people are now limited to 10Mbps or AnyPoint-like technologies. People who installed Cat5 are happy today. People who install Cat5 today will be happy today, but what about tomorrow?
There may be a day when some broadband provider drops off a >100Mbps cable to your curb for $50/month. There may be a day when home entertainment will take advantage of Gigabit Ethernet or Infiniband over twisted pair or whatever the next technology is. I recommend that you get spools of Avaya Gigabit now. The price difference is only about $50-$60 per spool over Cat5, and you're probably lumping this expense into a mortgage, so you're not likely to see the difference. Also, it assures you that you're putting in quality cabling. You have no idea what brand/quality of Cat5 cable your contractor is going to install.
Find yourself a good contractor. This contractor will have installed many homes before and might even have some active tract home contracts now that Cat5 is catching on. They'd not specialize in just Cat5, but also coax, telco, and security wiring. They would also make sure they install not just the cable, but the right jacks and wall plates at each location. I can't tell you how much time it wastes if you have to install your own jacks.
Have one set of twisted pair go mostly to a phone punch block (110 or 66, doesn't matter) and have the other be terminated (TIA-568A) into a TIA-568A 110-block with ethernet RJ-45 outlets.
Consider also instaling dual RG6 (coax cable) while you install your dual ethernets. Just like your ethernet an phone wiring, you'll want to centralize the place from where you want your lines run. You never know if/when your cable company will require two cables to get more channels. You never know if you or the next owner will become a satellite+cable channel junkie.
As for fiber, well, there aren't many proven applications for it, and it's more expensive, and there are fewer people out on the market who can do it right.
I think someone else mentioned using metal conduits around your wiring. I agree, though it might be pricey.
If your wiring is very important to you, and if you plan to spend more than 5 years in this house, the extra money you pay in your mortgage to get the job done right is well worth it. If you plan to flip the house after two years, just install two cat5 and one coax to each bedroom and each work/play area to be done with it.
Also, don't forget elsectrical outlets! It sucks to have all your ethernet bundles available in a space where you have no electrical outlets for your network gear and computers. Doh! The outlet near the central point of your wiring be better off with it's own 15-amp fuse.
-ez
PS: As an alternative, there's always wireless. Consider how/where you'd want to install wireless bridges. Consider the security implications, too.
I'm noticing >500ms ping times to a local router beyond my cable box - a symptom that I've seen happen before they disable modems. Upon calling Adelphia (Carlsbad, San Diego, CA), they said that they're sutting off today.
It should be coming back online on/around Dec 16th as a new local service.
- ez
(Just FYI)
Wasn't that the plotline for Dune? He who controls the Spice, controls the Universe.
- ez
Logging in via dialup becasue my ISP (Adelphia@Home) jumped the gun on disabling my modem this morning.
BTW: Noticed ping times go from 17ms to 300+ms last night. It shoulda been my early warning indicator.
So I went though all the domains, and noticed that only z.com seems to be a functioning website (redirects to Nissan's website.)
I was the first to register Z.COM. IANA once gave a directive that said, "all one-letter names shall be reserved to enable name-hashing at a later time". Working for a company that registered domain names on a daily basis, I thought, "If X.ORG can have a domain name, why can't I register Z.COM?" To my surprise, it worked! The following month, IANA gobbled up all the rest of the one-letter names.
A few years later, I started having people knock on my door monthly saying they'd buy or trade my domain. They didn't see much of a value to it, and neither did I. While I was a bit altruistic, I did have a price in mind where I'd do away with my domain. One day this guy offered me 50% more than that price, so I took it. It went toward a down payment on a house that later made me some real money.
The guy tried to make a simple Z.COM web portal out of it. Their gimmick was that all one had to do was hit "z" on their web browser address, and poof, there you were at Z.COM. The portal never gained momentum.
Other people bought it from him and tried again to make a portal out of it, but their gimmick was to give "lifetime" e-mail accounts if they visited the portal regularly. Again, another Z.COM portal failed, and those "lifetime" addresses disappeared with it.
The next purchaser was apparently IDEAlab. They never did anything with it and with their financial demise probably thought they should sell/dump it for whatever they could.
Enter Nissan. My guess is that they might release or re-release a "Z" car in the future.
I mildly regret selling the name away. I thought the purchaser would have done something better with it. I could give Nissan a web redirect as good as anyone else.
--
Eric Z iegast
eric@z.com
uunet!z!eric
Interestingly enough, NeuStar is headed by a bunch of folk that include an IETF dignitary or two (eg. Bill Manning).
The following document details how they're going to implement everything. Section B-3 goes into what you can and can't register under ".us" as well as gives info about the trademark "sunrise" period. At least it looks like they did their homework. There won't be any "com.us" or "xxx.us", but "screw.us" and "slashdot.us" seem quite feasible.
While I don't agree with them in principle, it looks like it could have been worse.
- ez
So many channels and still nothing worth watching.
Ditto (low cost, no GPL, publix fixes, stable).
Also, it's quick to implement. Any reasonable sysadmin can make a NAS server out of a machine and OS they already know. Installing and administering a NAS server is what the masses are trying to avoid when they buy an "appliance". In many cases, a NAS appliance is simply a server with a quick installation process and an easy to use configuration interface.
Here's what I think everyone uses...
Procom (started with BSD, FreeBSD?)
VA Linux (pre-implosion, based on Linux with ext3 & Mylex extensions)
Cobalt (Linux)
Sun (remember the Netra NFS server? Solaris)
Dell (Microsoft-powered NAS)
Compaq (Microsoft-powered NAS)
Maxtor (Microsoft-powered NAS)
Novell (yes, they have one, supposedly based on top of Netware-derived platform)
Netapp (custom OS, but suspect it was BSD-derived in early years)
EMC Clariion (VxWorks + Crosstor)
EMC Celerra (own RTOS + Linux control station)
MTI (older Crosstor on their own RTOS?)
Auspex (own RTOS[?] supported by Windows or SunOS control station)
AMI (FreeBSD)
... and countless others.
The underlying theme is "go with what you know". Each has their benefits and drawbacks. The companies have to support the products they create. What OS they know how to support probably influenced the underlying OS they used to implement their product.
AMI knows motherboards, BIOS, and RAID controllers. The software needed to make their hardware into a server was already available with FreeBSD and ported applications. Not that it's a trvial task, but all that they needed to add was a GUI, maybe a front panel, a simple installation process and some support for the product to keep their customers from having to "administer" a "server". Instead, the customers "configure" their "appliance".
The AMI products seem like good solutions for workgroups, but some caveats might apply:
- redundancy - seem to have little to no high availability features outside of the RAID controller
- heat - installing a few here and there are fine, but thinking you can easily put 320 drives in a 40U rack might not be prudent
- backup costs - Veritas NDMP slave licenses aren't cheap (but perhaps others are?)
- performance - there are higher-end products on the market for those that need more throughput or I/O operations per second
-ez
- If we've learned anything fom the Internet, it's that people will always find alternate paths around censorship. Some new USENET groups will inevitably pop up to replace the need of the former groups. For example, alt.binaries.penthouse could find a new home in alt.religion.kibology.priestess. Other distribution formats (eg: cache pre-population of encrypted content) could eventually replace USENET binary distribution.
- If Excite@Home had a spine, they might have phrased it like this...
Here's a post I made a long time ago that tells you a little bit about how alt.* newsgroups work... ALT FAQ- Due to our lack of operating capital and the fact that we really don't have the staff and resources to the administer stupid content that some of our users like to download, we're discontinuing support for the following newsgroups:
...etc...
Were they smart and also had some balls, they'd stop supporting USENET, period. Instead, they'd give each customer of both @Home and GigaNews a $5/month credit to use GigaNews. In return, GigaNews could pay @Home $XX,000/month for better bandwidth to the @Home user base. @Home would additionally free up expensive storage/servers/staff resources and be done with any liability with USENET.You can go somewhere like Giganews if you really see value in it.
Disclaimers: I'm a former Excite@Home employee. For that, you can label this post as a Troll. The advice, though, about USENET outsourcing is applicable to many ISPs. I have no affiliation at all with GigaNews.
`nuff said.
Eric Ziegast
Blue Mountain Arts
Personally, I like wooden mailboxes the best; you know, the ones that look like a little bird house?
Were I more cool, I'd put a little X.10 wireless sensor in it to let me know when it has been opened. I might link it to an e-mail/pager that says "You have mail!".
Just kidding. I've been using MH forever (late 80's). It's just so darn flexible for programmers to search their e-mail. I like my ability to search with Unix command line tools through zillions of messages I've sent/received to anyone. I've been using it since 1988. For MIME processing, mhn isn't that great, so I just bounce copies to Yahoo or Mail.com mailboxes so I can view them with my web browser du jour and then delete them. Instead of POP or IMAP, I suck my mailbox from a mail server using ssh and incorperate mail into MH on my local secure desktop.